argument to be having. i think we should talk about attentiveness of impressions and so, sorry, but a lot of web sites, essentially a digital yard sign. it's the same television yard sign during the daytime news, no one is watching it. but program, there's a place for that at a certain price but we should place a premium on advertising such as true view, people opt into and are paying attention to. i think that's why live sports is more important on television by far. video on demand with the cable, essentially true view on television. and the final piece is holistic media plans. instead of saying the media budget is x, and this percentage goes to tv and this percentage goes to digital. let's allocate to the thing that the medium is best at. broadcast is best at generating a lot of reach, targeting people at once. but the price efficiency and ability to scale down to reach frequency to your target yenls

as tremendous financial diminishment returns. 500 broadcast points that we would get approximately five impressions through our target audience but then we would use cable and a tremendous amount of target digital to the same audience that we were getting closer to 45 or 50 impressions of total message penetration before we would change traffic. that situation didn't play out in every instance but the guiding prinls ciple was arbitr true viewpoint but rather a holistic delivery of impressions of the audience. and final, creative consistency. everyone understands this principle. everyone understands the idea of having a theme. everyone understands the value of kark terse acharacters and s probably rolling eyes a bit but we don't do this often in politics. up front planning, commitment to something in particular and i think that in order to cut

through, this is what we're competing against more than the other side and stick to this as often as possible. two states proud of this. pennsylvania, with this bike messenger who carried almost all of our spots, with a complicated message and he was able to hold people's attention long enough to deliver a complex message effectively. we had another example, i don't have it up here, but arguably more creative example in ohio with a quote unquote fake newscast. we had an actress, several states show all the places. so this idea of having a character and creating a consistent theme and a consistent brand across all of her digital and television advertising helps you deliver the message. proof is in the pudding. senate majority pack, own bike messenger talking about how our bike messenger was evil but that's how you know it worked. and the whole thing was shady

katy mcginnty and a week after the election, the student body was trolling the other student body using the shady katy tag line because that was her alumni. and creative skinconsistency, s states, this is the color, fonts, look and feel up to, here's our geico gecko. it takes a lot of early planning, a lot of committing to the idea by the whole team and a tremendous amount of research. look, i think we all agree on the problems. i think we all agree on the solutions. the devil is in the details and committing to research and budgets up front, i think, it's the name of the game. so thank you. [ applause ] >> thank you. next panel up here with andrea, betsy, chase, michael, peter.

while on stage, this was the first year we saw political ads hit youtube leader board. can you pause this one second? so political ads are known for being really bad con tenlt. youtube leader board focuses each month on the top ten performing ads online. whether they're organic or have some advertising behind it. this year, this cycle, we had 5 political ads hit the youtube leader boards. i'll show some of them to you today. we kicked off with the closing ad from rump atrump and another presidential candidate and then three ads from the senate campaigns and then a theme here that daniel just hit on. so i'm going to let y'all see if you can figure out the three from the ads shown after this one. >> i understand that when mainstream media covers immigration, it doesn't often see it as an economic issue but

i can tell you, it's a very personal economic issue. and i will say the politics of it would be very, very different if a bunch of lawyers and bankers were crossing the rio grand or journalism degrees were driving down the wages in the press, then we would see stories about the economic calamity that is befalling our nation. if i'm elected president, we will triple the border patrol, we will build a wall that works, we will secure the border. i'm ted cruz and i approve this message. ♪ oh say can you see ♪ by the dawn's early light >> holly was 21 when she died.

we lose 129 kids a day to heroin and the only person that i've seen standing up there screaming almost daily is senator portman. he gives as much time and energy and love to this as any of us parents who have lost. he truly listens and tries to implement plans to make a difference. c.a.r. e. is such an important step in what's happening. i know it will save lives. >> throughout his career, john mccain has been a true friend to the hispanic community. he has fought for comprehensive immigration reform and for our small businesses. i urge you to vote on november

8th and i urge you to vote from my friend, john mccain. >> my name is nicole craig. >> i'm kevin craig. >> we live in green bay. we came home with grace. it was a nightmare we wouldn't wake up from. 25 children waiting to come home. there wasn't clear reasons why. the adoptions were completed. they were final. we had to go to washington. we had a strong player in senator johnson. he's a father and a grandfather and he was going to do whatever he could to get these kids home. dear nicole, thank you for visiting and sharing your story. hopefully, we can find a solution so you can bring elizabeth grace home. your family are in our thoughts and prayers, sincerely, ron. i can't tell you how that felt. i don't think i believed it. and then having senator johnson show up, that was just the icing on the cake for us because i knew at that point it was as

important for him to come home as it was to us. >> i'm ron johnson and i approve this message. >> come on up here. hopefully we'll have everyone talk about their ads. >> managed the wisconsin senate campaign that took us all by surprise this fall. ron johnson pulled up a come from behind victory after they had pulled their advertising. chase campbell, a digital advertising specialist with harris media whose resume includes two races, could have, should have been on the races in retrospect but hardly think of him that way because of mitch

mcconnell and rob portman in 2016. michael duncan, a digital strategist with cavalry. he works with a couple now and races like indiana senator-elect todd young and john mccain's reelection and mike lee and ted cruz for freedom works. and peter who managed another one of the most uphill races of the 2016 cycle for pat toomey. we want to first talk about the ads we just saw. >> so the grace ad that you just saw was something that in the work for a long time. we did research early on identifying the constituents in the story and we knew we were going to be outspent so we knew that we needed to be and we knew

they were going to see if they would only care about billionaires. and we did our research early and basically to go around and make sure we captured constituent stories like this one. so this was a spot that did ultimately end up on television but the intent, and it was powerful and so we decided to run wit and ultimately, we had a two minute video, 60 second video and a 30 second video and then ultimately, ran it on tv and we did have it running towards women on youtube on a low burn from, you know, the minute it went up through election day.

and we saw with the survey those who watched it two times more likely to vote for ron johnson. so it was a great story to tell. it served a purpose and something we were proud of. >> the follow-up, republicans talked about a shift in strategy and things that were shifting resources around. is that part of that? >> we knew we were going to be out raised and outspent. this was part of the postlabor day reset that we did.

once folks in wisconsin were not enjoying the summertime anymore and kind of paying taex, we hit the ground running but we had laid a lot of that groundwork early to make sure that we were ready to do that. >> heard this before, but about this time in 2014, an executive director for reelection in 2016 and showed them a slide show since the last time in 2010 and told them the most successful campaigns of 2014 spent five times as much as the average 12 campaigns. you were part of the senate campaigns including an unlikely early adopter in richard shelby but we'll get back to that later. can you tell us what that means? what aspects of a modern campaign are covered under digital?

>> you know, i think if you go back to like 2010, people saw digitals part of the comms department. we were the people that took the press release and posted it on facebook or on twitter and like that was it. it was like, that's a victory. and then people started to realize the potential for digital to raise money and so then it became part of the fund raising operation. and people started to realize that the way mied ya consumption, and that was a huge part of the advertising budget and because of the part of the budget, part of the data so

digital went from being on the periphery as everything and richard shelby who we worked on in his primary and todd young or mccain or portman, or ron johnson or toomey at the ground level. i used to work at harris media with my buddy and called me on christmas day to hire us because it was that important to have that part of the campaign locked up two years for his election. not many recognize how critical

the digital component of the campaign and starting early kind of gives you that leg up. >> they will probably need to get on it right mow. how did you guys do that and what did you learn from the ohio race? began building his online audience and remail list and controlling the message on searches as soon as we could begin spending money. and his, i think, coming off of 2014 and later mcconnell had in his race, he started early as

well and began doing stuff with him in march of 2013 for that race. really kind of set the framework for 2016 candidates and senator portman, we were testing video messages from him in the beginning of the summer, so a lot of, freblgs, the holiday, we just saw, we began running that and tests online at the end of april and beginning of may far before it came on to television. and a leader on, and it was more of a reminder an we did use some cool tools that it had used before with the great team at google. talking about display advertising and we did a test actually because it's a common

conception in digital advertising junk. we don't know who has seen it. it might be a computer. not even a person. that's something we wanted to test with google. so we ran a display campaign for a week on the c.a.r.e. legislation related back to the video and all we said is look at what senator portman has done to curb heroin addiction in ohio. and we ran after his target audience of likely republican voters and for the two weeks after that, people exposed to the impression, not people who clicked on it but just people who saw the ad somewhere online went and searched for the senate 600% more often than someone in the audience who never has seen the ad and sort of searching there too. that was the cool tool we used to help bring some validity back to the display advertisement that we were running for the senator. >> and in laparticular, that wa

on earned media hits with heroin stories right, as well? you matched your ads to the earned media? >> there was a combination of targeted advertising to the ohioans likely to be moved on the heroin issue. >> peter, something that senator toomey talked about earlier on, as a republican in pennsylvania, was his own race independent of the presidential campaign no matter who the nominee was and in the end on election night, we saw interesting results when you compare toomey's counties to trumps. did that surprise you that you won different voters than trump did and does that sort of speak to the way senate campaigns can run more independently than we thought from presidential race? >> we knew, we always knew pennsylvania was going to be an uphill climb no matter who the presidential nominee was. so he planned to always run

ahead of them. what is amazing in pennsylvania, you have two statewide republicans win and their maps are two totally. they're very different from one another. president-elect trump found a path through the rest of the state outside of really the philly media market that we just didn't know if that path was available to us. and i think we went in and we knew we were going to have to find a large number of split ticket voters and go after them, identify what issues mattered to them and then talk to them across all platforms and really feed them information that they would care about and i think we did that, obviously, we did that effectively. i think the philly media market alone, 100 or 120,000 split ticket voters itself and just the philly media market alone and we won statewide by just under 100,000 votes. so i think it is amazing the two different paths, but, so we were

surprised on election night when that all kind of came about. >> the beginning of this cycle, all we heard is that senate races can run only so far from a presidential race and the cycle every race went the same as the presidential. does that mean there's potential for a senate race to turn its own course especially with the new technology you were using this time around? >> we invested heavily in, you know, using data and digital to target people and to really brand the senator on a good public servant. i mean, ultimately, you, as an incouple benlkucoup incumbent, have a anythinegativ. and they'll put it in that

light. and worked to really save this little girl's life and getting the rules saved to get her lung transfer and we ran an ad on that in the philly media market. we talked about the senator's independence on the toomey mansion legislation. a lot. we talked about the giffords endorse. and talked about the senators, really, his record and who he was and how, really, the work he had done. we spent the entire primary talking and into the general talking about the jobs he had brought back to pennsylvania. we ran targeted mail around the state and in southeastern pennsylvania. we talked about the refineries that he worked to saved. the 9/11 air wing outside of pittsburgh. we sent mail to all the surrounding towns talking about that work as all the jobs related to that and then on digital, we matched all of those same universes so the mail piece

to drop on a monday, digital on a wednesday to the share the same messaging. we were, it was all about repetition and then taking, highlighting the work that the senator did and really talking about who he was and getting people to vote for the senator and really his work as a legislator and that he deserved another six years to go back to washington. >> and this question is kind of for all of you guys. i think the portman campaign would say all of the cool tools that you have and all of the cool media you're using is without volunteers on the ground gathering information about the voters that you're delivering it to. is that a really big part of the campaign and did you do anything similar with the portman campaign that recruited like 500 volunteers? were you guys doing something similar? >> yeah, for senator portman specifically, the field operation from the campaign was accessing and inside the same database we used on the digital

side. they had all the same information that we had every single day and it was a real key component to his victory because everyone was working off the same information and everyone could speak to this. sometimes, the field likes to use their own tool and fundraiser likes to use their own donation processer and digital person likes to use their own e-mail platform and it's all working on different things and that was never the case on senator portman's campaign and i think that actually not looking at digital as just a piece but something that the entire campaign can be a part of and including every piece in it i think is a real key to success in today's digital world. >> same thing with us. we had a fantastic in-house data director who, you know, made sure that everything that we were doing whether it was the doors we knocked on, the calls we made. the digital universes we were targeting. our tv, everything, mail pieces,

everything was from the same song book and it did make a difference. for us, we had a limited budget and we did have to be smart and efficient about everything that we did. and, you know, we were able to track, you know, every target effort that we had so that we knew, you know, like where there was movement and where we needed to supplement and where we felt good about things and it did make a huge difference in terms of our efficiency. >> and connecting all of those, sorry, did you have? >> i was just going to add something. the mccain race, this is an interesting anecdote in the primary, you know, john mccain actually lost election day by 10 points. but because there's a huge drive in arizona, like with the mai

mail-in ballots except it's like 80% of the e llectorate will eay vote. so we have this permanent early voter list that we were targeting online in the 26, 27 days before election day, all of the voter targeting. anywhere that we could possibly match somebody or ip target somebody or just match registration data on facebook or like mobile numbers on twitter. anything we could do to maximize the reach and frequency to that permanent early voter list was like absolutely key to winning that primary and all of that data utilized by the team to call through and change it because if we could find somebody online and say, oh, by the way, that ballot is on your kitchen table, please mail it in and kirkpatrick goes on television and kelly ward gets $200,000 to put in the last ten days and so basically using the

digital as something that could feed into the rest of that field operation that was like the number one thing that was useful there. >> and betsy, this is something i've within sobeen super curiou. you worked with the scott walker campaign. wondering if you have a strong digital campaign used to help another campaign down the road. did you find the information the walker folks had was helpful the you and do you think the guys you had was helpful to other campaigns? >> yeah, absolutely. i see a huge part of that effort here, so thank you. yeah, i mean, i think, you know, it just continues to be built upon and to get better and be refined. we started with our target universe looking at what we called the walker johnston gap. we used the walker post 2014 as the high water mark and built universes back from there and i think that, you know, we want to make sure that that's available and it all lives in the state

party and whoever runs in 2018 against tammy baldwin in wisconsin will have that available to them. it's just something that continues to be refined and made better with every point of contact that we had, every volunteer knock, every digital ad, every touch across any medium that's made refines the file and can be used down the road. >> owned by the state party or travel with the campaign at all? does rob portman need to wield this book to josh mandel? >> we'll see if there's a primary in ohio, but i can tell you senator portman is not stopping that election day is coming on. he's going to continue to keep his file active and make sure that the people in his database are being communicated with on a regular basis so that whoever is the eventual republican nominee for governor, senator, down the line has a good file from which to work.

>> this was particularly interesting to me this cycle the way you guys got involved in primary. has digital advertising changed the way with a name id? >> a couple of things on primaries. that's a really good point and not one that i think was lost on, you know, the leadership on our side and the people that run senate races. i think the whole attitude has changed. you know, a few cycles ago it was, we're not going to talk about the primary opponent with low name id because we don't want to raise their awareness. but what we've seen and having worked on both sides of this coin, right, like having worked in primaries or insurgent care candidates, you see the problem in republicans is online. those voters overindex for usage

of the sberninternet. t the, if you are not defining your opponent before they have a chance and someone says shock poll three weeks before the election and suddenly, you're neck and neck with this guy with 20%. you want to get them under water. and early to those people who know who they are and the way you do that is online. you know, i think what we see in the primaries is the digital shows in the verbatim statements you get read back to you from the people that you're calling. and so if you use digital to really change that sort of flow. we talk a lot about fake news now after the election. and i mean, honestly if you worked in the republican primary, you've been dealing

with this problem like forever, right? if you look at what people say about your candidate, you've dealt with the issue of fake news before. but like now that it's affecting the general election, now suddenly, like the liberals cares and the media cares and everyone cares, right? and if you worked in the senate office this 2010, 2012, you've dealt with this problem literally and what it is is info flow problem. it's not your fave on fave. it's literally what people are reading every day about your candidate. and they're reading it on facebook or they're seeing it on twitter or basically what i'm saying is your number one task in these primaries is to change that info flow, is to get people to read your content and not what you would deem to be fake news or biassed news or what some radio host is saying because you took some vote for cloture that wasn't the correct vote. if you're going to have an

insurgent candidate against you in the republican primary, if you're not spending money online, you will lose. >> a quick lightning round? since we listen to brad talk about what the trump campaign did, and you're all professionals, anything you want to take forward going to senate campaign? >> i think that, you know, their use of earned media with everything that they did is something that should be replicated. i think digital ads are treated like tv ads now. and it's a lesson that the trump team did really well and i think we can all use to our advantage, particularly, when you've got limited budgets, you know, treating your digital ads like a tv ad and making sure that you're raising money off of them, making sure you're putting on and pitching stories, it's the greatest thing that just happened is really importantly and only expands your reach.

so i think that's something that the trump team did very well and we can all take from. >> i can only hope to work with a candidate in the future that carries earned media and news cycle like brad had the opportunity to do. but i think of all the things he said, one of it was the seamless hierarc hierarchy. that wasn't a lot of hands in the cookie jar when it came to putting it out. a lot heard the stories about 2012 romney, 100 people had to look at a tweet before it went out. that's an exaggeration but it was a lot. we say that after elections, joke about it and then we forget about it and then we go to the next campaign and some of the same problems come back again. so i think, you know, we still have candidates today that want to look at every e-mail before it goes out and change one word or, you know, and that's a problem. i think as much there's a seamless hierarchy there for decision making, it's better for all campaigns. >> i really liked the square

video for fund raising on facebook. i thought that was very unique. i liked the native use of subtitles. i think that's very good in a way that draws people's attention. it's not just to exist on facebook. we immedianeed to not just say and we're the best, but like think about ways like that that can sort of disrupt these other advertising venues or other social media platforms to really engage with the user. not just take the 30 second spot and run it. >> i'll go two things. one is, i think, brad's ability to understand donald trump's voice is actually critical to what they were able to get done and finding, building a team on a campaign, especially a team putting out public information that understands who your candidate is and what, how they

speak and being able to kind of capture that helps to streamline that process and two is the social media following. i will be completely honest, i didn't totally buy into it when i started. my job in twin. i knew there was benefits down the road, but it was like, we're just growing our facebook likes and always kind of asking, what's the point? but really what i found is the ability to turn people out to events was cheap. i mean, $5 cheap that you could be able to target. you could, through social media, to be able to get people to come to an event and even a center race where it's much harder. we didn't get the size of crowds that mr. trump got but to get 100 people in a county with joni ernst, it helps in ways you

don't see when you're getting up and running but in the end, it does give you benefits to be able to talk your people and get them involved. >> as somebody who covered these races for the past two years, it's a privilege to have all of you on the stage. cool things happening in all the races. >> thank you. [ applause ] okay, next up, ted peterson.

>> hi. what's up? all right, rolling. i'm ted peterson. i was digital director for the nrcc independent expenditure. played in 29 races defending the republican's largest majority in 80 years. the media expected us to lose around 15 going into election day and we lost 6. the nrccie spent digital out of the $75 million budget. 30% of the total media spend. digital budget increased by 60%. for the first time, budgeted for digital creative video. i want to touch briefly on two

google relevant topics from ie. search strategy and 15 videos and bumpers. start with search. okay. so when a voter and third district searched for jim mower on google. a negative ad. if they clicked on the voter guide ad, this is where they landed. you'll learn more about jim mowrer but kind of negative. you'll notice paid for by nrcc. zen search was the first thing to go for it. a little real clear politics. it's a site with information about all of our candidates,

content about republican candidates was positive and content about democratic candidates was negative. you'll note the side bar and these were linked to the store. we wanted to give a news feel and tailored content for the medium. we created a site that didn't look like an acttack ad but had the attack hits. over 25,000 from the search ads and that's great about these visitors, the search ads, these were individuals actively seeking information about democratic candidates in the target district and we were able to reach them through the voter guide. back to the search results. another nrcc ad on this page. it's the wrong for us ad and when voters click here, they're directed to the 32nd tv spot rung on tv and elsewhere on the

internet. the wrong for us.com landing pages is where we directed users to pick on display ads and designed to look similar so every page looked different. you can watch the 30 second video and this static image doesn't do it justice but you could scroll down, reach other videos and sign up for more information. back to the search ads. one thing we did a little differently this cycle was bidding on the same search term with two ads. some argue that we were stupid to bid against ourselves and i note our buyers did implement a strategy we weren't bidding too much against ourselves but you werely made an effort this cycle to own search. we owned the democrats names and the d trip was nowhere to be found and when anyone searched for a democrat in one of our target districts, we had two

totally different creative styles and deliver the message to them. i think we benefitted from owning the top two search results. let's move on to the digital first creative. most of you probably saw this running in the dc market. if i could get it to play. is there a play button back there? oh, it's not going to be able to play? okay. all right. well, anyway, so everybody saw this. i can improvise. luann bennett. owned a parking lot. preschool, yada yada. 30 second ad, everybody saw it. what we also did for almost every tv ad is break it down to 15 seconds. we had this 15 second version you wouldn't have seen on your tv but would have seen it because we're running it on

youtube, targeting it to you and i can't play it. but it has the same message. and for some districts, we went further and developed youtube 6 second bumper ads and this was just a 6 second ad. a small clip from the tv creative. the 30 second ad broadcasted and then the 15 and 6 second and i think it was a big part of our success on election day. and one last thing to leave you with, i think we were particularly successful this cycle because we started with a plan. we planned to tailor content for the medium with digital first creative, $8 million on digital issue built out and stuck to the plan. thank you. >> next up, a panel on ballot

props. i'm going to welcome amanda, chris georgia and lauren benson. there were more ballot props this year that we worked with than ever before and i'm sure you'll all start seeing the same. >> there we go. my name is lauren bunn sson. i am at google. i am joined by amanda bloom, director at basque as well as chris georgia at fp 1 strategy so thank you for joining me but before we jump into our panel, i have a very exciting topic. which i know we're very close to drinks, so we'll go as fast as we can through the digital

research ad effectiveness results on ballot props, so don't get too excited. thanks. so in my role at google, i have the job of focusing on voters. what are voters doing on digital? and the way we understand voters and their intent in election cycles is looking at multiple different data points. everything from what they're typing in the google search boxes, they're typing very interesting things to partnering with third party research firms. so today, we'll walk through our evolvement of research and findings on ballot props this cycle. but at the end of every cycle, we seem to be in this position. questioning the effectiveness of advertising at the overall level, all the way down to digital and if i had a nickel for every time i was asked in the research industry to show me the effectiveness of digital

ads, i might be the next billionaire running for president. this is a number one question asked we get. how we set up research this cycle was to focus on three different buckets. really focus on the shift that we're seeing in the brand in consumer world of the time spent on media and 2 hours online to every 1 hour on tv and really where it gets interesting is to go into the influence in the way of digital and finally, rounding out the effectiveness but google, we don't want to stop. if ads are working, we want to go beyond to understand how, where, and for whom the ads are working. so we took an innovative approach this year and i wouldn't be on stage without the help of chris and amanda and their partnership.

so we worked with come score and working with their national panel but we know elections are won and lost at the local level and we thought there was a huge opportunity to understand more and looked at 35 states and had com scores cut their panel and screened in for likely voters. and the third stop is tagging the creative to stay with a significant lift and the true controlled and exposed experiment if digital ads worked. we'll talk about the strategies and how chris and amanda executed these strategies and pivoted throughout the cycle but we have some results. so we look at all if

persuadability. knowledgeability, intent to vote and some intent to recall. we saw significant lift in the double digit percentage points across multiple metrics. more to come and reaching out to your google team, this is scratching the surface. but i want to start with chris and amanda. so the first question that we had is as we've heard a lot today what made a successful digital campaign on the senate presidential level, i would love to hear from your perspective working on ballot props as you worked on ballot props, what were some of the challenges and differences executing a ballot prop strategy? >> okay.

>> there's no, there's very little awareness going into all of this. in california, for example, there were 17 ballot initiatives. and the voter guide was 250 pages long. that's just the statewide voter guide. that doesn't include your local races and initiatives. so with we had to battle and the initiatives is in general awareness does lead to favorability but we do have that in our favor. >> i think that the audience side of that is really

republican voters and figure out who that audience is and across party lines. with that and enough with the effort, have a great investment. something they had on our end that we were able to take microtargeting data, combine our online targeting, not only with what was in the mail but what we were doing on tv as well. we partnered with analytics, did extra work with them and were able to target online. and so all of that kind of fueled by if fact that we had to go find this audience. >> great. so chris, you came from behind in double digits and ended up meeting and then exceeding by, we think, 70 points. >> it was a good night. >> it was a great night. one of the reasons that happened

is because it wasn't just a well funded effort. we ended up with the ballot proposition and new jersey is in and new york is the most expensive. you're talking about for outside groups, $5,000 a point. so to go in and play there, it's incredibly expensive. we had well funded effort and because of that well funded effort, we were able to start early on in the process. we had researched starting back in may before it was decided what number on the ballot. this was going to be yet and then once we decided, we went big and never came down. >> great. amanda, with this research of what did you find was the most surprising or any pivoting from your internal research? is that what was coming in as you moved towards election day?

so one thing, we're in a similar boat with one of these campaigns. down by 25 points and ended up 4. either on the internet or no tv content whatsoever. that was huge. one thing came from the study we did was that and there was daily polling going on with these campaigns too like who we need to target and how we need to change. but with the study, the thing that surprised me the most is that we all believe that digital can move voters. we believe that it was nice to see it in numbers and present that to our clients. but the thing that i really enjoyed seeing was as we changed messaging, the lift in message recall popped and we could, i was seeing on the results that

the messaging that we are running in september was popping and then kind of integrating our messaging into that but it wasn't, you really didn't see it move and basically found that te message that we were weaving throughout was really from polling and from focus groups was the message that was going to win us this campaign. so we went full bore the last couple weeks of the election. cross stream tv, digital radio, everything digital, direct mail, everything. i mean, everything had the exact same imagery in it. all across the board. and we saw a lift of, i think 32 points or 25 points, i can't remember exactly what the number was, but it was huge. so to be able to see the difference in just message recall as we were changing the messaging, it actually shows that it's working. >> so -- like you said, the

numbers might nobody be assigned until later. so what tools from a digital perspective did you use to educate voters, as this was pretty different from the awareness levels that we were working on presidential incentive. >> on our end, it was particularly challenging because ours was about casinos in a state where you can gamble. so we had to go in and do an education component of this campaign that made our search strategy challenging. people much starter than me on our team worked with google to figure out how to make that buy. it was something we were constantly dealing with and bidding against, the same folks we were trying to stop. >> we went up early on with advertising before we even had a ballot number and because if anybody was hearing about something to do with it, we were basically using terms like if they're searching for something even loosely related to this, let's be able to get our message

in front of them. so search was extremely important component of all of that. another thing with how confusing all of this was is that we had to figure out how to word our strategy appropriately. so for one ballot initiative they were running, it was -- we were the no side and the initiative was basically to fund education, and we knew that people in that state supported more funding for education. everybody thought education needed more funding, so we couldn't go at this from, no the state doesn't need more funding for education, so we had to figure out how to word that appropriately. and that all translated into the search ads we were doing and we went up for most of the campaigns in early summer and went up hard and didn't come down. >> i would say another component of this is frequency. it's something that in dealing with an education campaign, you need to get your frequency up in

order to burn that message effectively. especially when there is not a face or can nate or media that will go along with it. we did two things along those lines. one of which we figured out cross, medium, what individual voters were likely to see on frequency. where we could, we augmented that with individual. tv is example of that. making sure the frequency was higher amongst the cord cutters, we knew the broadcast buy was less likely to hit them. then the other thing that we were able to do is we were able to download through data transfer 2.0. the cookie device and d log of frequency on a daily basis and match that up. we could go in and identify folks that were below the frequency we needed to hit them and increase that as we saw fit. >> i think we have time just for one more question. so what do you both think is on

the horizon, we pushed some boundaries of even pushing com score to look at state level panels and had some funny think kind of pushing the industry forward and operating in this type of research. but what do you think is next or where would you like to see us go with the digital research frontier? >> it was really disappointing to me that com score could not measure the mobile impact of what we were doing. with the shifting trends toward mobile consumption, most of the revenue is coming from mobile devices, we couldn't measure what the lift was from that mobile component of it. i would really like to see that get pushed forward. in addition, it was hard to get significant size and we were operating one of the largest geographic dmas in the country. the research has a far amount to go even though the targeting is certainly where it needs to be. >> yeah, we were concerned about the size as well and -- but i think from a research

standpoint, it's -- it's hard to say. one thing -- this is kind of loosely related that we found with a few of the ballot initiatives that we were running, from polling numbers, was that 20 to 25% of our voters, i mentioned this earlier, were not able to be reached with tv advertising and so i'm not sure how many of those people are actually measured through the com score study because the panel was relatively small. it would be interesting to see the difference in impact with those individuals specifically versus just everybody else out there on the internet who could have potentially been getting the tv ad on top of the digital advertising. >> thank you, both. [ applause ] >> thank you. matt l ira you are up next. thank you for dealing with our av problems.

and for your pleasure, matt has decided to go ala tim cook, no slides, just tim. >> no slides, i don't like the accountability of slides. i told you. but before i start kind of what i was planning on talking about, i want to share something that struck me as i was in the back and listening to the panels and all the earlier discussions, that's the assemblage of people in this room. i see digital strategist, creatives, people who work on committeeses, candidates and causes and people worked their way up from the bottom and people come in from the top and make a difference. all of you have been fires for the fact that digital is changing the way that campaigns are being run. and whether it's local initiative or the presidency of the united states, you've all been willing to take the risk, put your reputations on the line and to fight for this new and better way of doing things. so i think as we're kind of

collectively maybe give ourselves a round of applause or what we've achieved and what is ahead. give yourselves a round of applause. [ applause ] >> obviously ward and brad, todd and shields and brian, it's a really great group of people that have really fought and stood up and put the reputation down when the chips needed to go down to make that change. now we have the opportunity to change the country. to my real talk, essentially we're seeing a change in media, this disrupted change in marketing and sell products and entertainment, politics, et cetera. and it's challenging. i think it's something that our democracy has seen before. you know, if you look back to abraham lincoln he said once that matthew brady and the cooper union speech made me president. matthew brady was a photographer in the 1850s and '60s, believe

it or not, that the time still photography was this disruptive new medium. they thought still photography was kid stuff, like ent tank distraction you throw to the side. and matthew brady and lincoln were able to realize that this is how you could shape the opinion of a back country lawyer and turn him into a national leader. and he leveraged that to lead a country through some of the most divisive times we've ever confronted. you zip ahead a few decades and ee fdr. he realized radio, this new disruptive media, could be used to communicate with people in the living rooms and forge a connection with the american people and guide them through a depression and world war. you zip forward to the rise of television, jfk has the audacity to do something weird and unusual and wear stage makeup to be on tv. the serious politicians wouldn't necessarily wanted to do that. he realized that the master a

new medium meant to do something a little weird at first but is core to what makes that medium distinct and different from its predecessor. one of the things that strikes me how digital is different is its interactivity. it's still going to have great sound, still has to have great audio and great imagine, but it has to be fundamentally interactive and strike people as an emotional connection through kind of interactive medium. so many people did that really well this cycle. so we're at this peak moment where we have the opportunity to run you know, the country with the house, the senate, and the white house, but i think we need to challenge ourselves because the media is continuing to change, digital will continue to rise, and we need to continue to find new ways to leverage disruptive thinking to make interactivity more central to how we run campaigns, how we

sell policies and ultimately how we govern. as a wrap up thought. one of my favorite economic theories is the sailing ship phenomenon, which is this idea if you look back at mid 19th century, there were four or five major ship builders and they had been dominant for centuries, by the end of the century not a since he will one of them existed. why, when faced with the self evident truth that steam technology was going to fundamentally build disrupt building the ships, why don't they start building steam and they were unable to make the leap that the world was changing. it's not an isolated story. if you look at why didn't xerox embrace the graphic user int ter face, why don't block bus buster

interfa interface streaming. if you don't disrupt, you will be disrupted. i think as the leaders of the conservative and republican community, at least as it relates to campaigns and campaign strategy in this room, we have an obligation not to rest on our law relevanurels. so thank you, and appreciate the opportunity [ applause ] >> thank you, now we're going to have the war room for san antonio from the trump campaign. we have jonathan swan, garrett lancing, garrett coby, brad pasquale, matt. molly schweickert, and i got to say this is a little bit better looking an than the war room in san antonio if you have been

there and can drinks are flowing. if you didn't have a mike -- >> well, this is a fantastic panel, i'm sure most of you are familiar with these people, this really is the donald trump digital team. these guys were the core of the team. jared kushner obviously approving things but these guys really were in different respects responsible for the operation. and you know, we've seen so much reporting, some of it correct, some of it slightly incorrect about the various roles. just slightly. so i'm keen to dig into that. but one thing i think everyone would be really interested to

hear and maybe we'll start with the rnc folks here, is can you just give us some insight into what it was like to walk into the donald trump operation when you first did, how did -- how much did you resemble a normal digital operation, was it just something completely different. just give us a bit of a flavor of it. >>. [ in audible ] >> so he had a business he started running, a beautiful office, but as he kind of shared for that 1.0 window, he was running a solo operation. you know he would pull in guys here and there. he had some folks i know he had to pull in, i'm sure at times. but it was, you know, something he was running solo. it was pretty impressive they

got so far in that way. and our first meeting, it was kind of a little rough around the edges i would say. we're like these rnc guys showing up down in texas. i'm saying so through that, we hung out there a couple days, came down a couple times, grew a really good relationship just working together. >> i think it was pretty sparse when we got down there. i think therefore surprising, but we had a good connection. everyone wanted to win. i think from our perspective, i knew coming in to the rnc that there was 16 or 17 campaigns and none of them therefore because of the resource split was really going to be able to have a digital operation that was -- could scale into general election size very well.

i think he had a well sized and scaled operation that could have done it, but we knew back in june or assumed back in june of '15 that wasn't going to be the case. i wanted the rnc operation to be as large as possible and have major e-mail operation list and staff ready to go at a moment's notice when the nominee walked in the door. so it was ironic or maybe perfect that the campaign that came out of the primary had the smallest digital operation. in terms of just man power, not footprint. but it ended up being a great match that we sewed together over the first few weeks and the rest is history. >> this one is for brad. i want to know to the extent to which you think not having the hillary clinton super structure was an advantage, so actually being smaller, not having as someone said 42 lawyers to sign

off on a tweet. do you think there were advantages as well as disadvantages to that? >> well, it was cheaper, one point it was all mr. trump's money, so it kept me around longer. i mean, so you know, i think we only spent $2.4 million until the convention total on digital and i would say $2.2 million, $3 million was probably all in ad buys. we built the website, the digital structure and everything i billed less than 30 grand, at a point when we were in iowa, that's not very much money. we built a store, built all the pieces, done everything, but again, i did that from the house. so -- >> what about in terms of function alt. >> functional? i think it's hard to look at version one and version two and version three that way. if you look at the end, i think our digital operation, by that

point, i can't say our scale was actually smaller. i might say the scale was larger. i don't know exactly how many people had. we had a pretty big -- we were a hundred plus people by the end. we had a lot of different assets there. it's unfair to look at one point of it, that had nothing to do with hillary, that would be a question for the other candidate and i think i had a heck of a primary candidate on my side there. it wasn't as much about us. you look at the general, we put pieces together fast, in partnership with the gop and rnc, cambridge, other partners, i was able to pull out -- my company is not small, they make it sound like i'm sitting there by myself, it's 70 people, it's a whole city block, it's not like it's a little building. i just hadn't pulled resources because i had to bill for those. by the campaign we started

pulling more and more resources, we outgrew our office and had to move to another location in the city and then we got almost a whole floor of that building and then i forced a lot of them to fly here because i didn't like flying to d.c., well san antonio, so they all, which wasn't a bad place in the united states to be stuck, it's sunny, no direct flights, but we can fix that now. >> it's good to see the power is not going to your head. can we send the mike down the other end. so one thing i think that a lot of people are curious about, i was talking to some republican operatives this morning and saying what would you be interested in knowing about. they want to know to what extent in your digital operation was persuasion and what extent was fund-raising, can you tell us about how both of those elements worked? >> one of the cambridge people, maybe, i have not spoken a lot.

>> obviously as brad talked about the -- that was about infrastructure, getting both new e-mails addresses to turn over to the e-mail marketing team and direct fund-raising, which we had a tremendous amount of success doing. as we came down into the cycle and understand the pocket of voters and understand mr. trump's story from an -- and identifying our supporters and making sure they had the tools to request ballots and understanding where to go on voting day, thez were important parts as well. it was digital first. whatever the various objectives of the campaign were, whether that was raising money or getting supporters to come out, digital tools were used in order to accomplish that. >> you want me to answer that. >> many wise, gary by far got most of it, the fund-raising.

the gop and rnc handled all of the digital fund-raising for the most part in partnership with the campaign and jiles. we would produce content, they would and gary would lead that team and then we had partners from third party companies to push the content. that was the majority of the money, i think, i don't know who how many millions of that is, i would say 70% rang. then we had a persuasion budget and gotv budget. that was mainly run again through the rnc through another portion of the operation that garrett ran. i don't think we did that much gotv digital through the campaign itself and then we produced content and then worked with them and then you had the persuasion budget which i believe was, i don't know, 10% to 15%, 20%. i think we ran our fund-raising effort in a way to be persuasive

as well. maybe gary can talk about that, i think that's a big difference of the way we did fund-raising to meet two goals at one time. >> i think first it's the way we're running the fund-raising in terms of our turn. we were aggressive and had a mindset of if we're getting three extra we're not doing it right we're leaving money on the table. we had such a short window to operate, we were bringing as many donors and dollars as possible, really the goal was when we were getting great return. roi, a third roi at times, we really kind of ramp up our volume to push it down to bring in more total dollars and if we're operating at 1.3 x that means we're doing better than in my mind than 3 x because 3 x we're leaving money on the table. in terms of how the operation probably helped with the persuasion, you know, trump followers got amazing kind of

engagement, so a lot of our content gotten gaugement, lot of our content got shared, we were extremely aggressive with testing and pushing envelope, what kind of messages worked for dr. and we found, you know, sometimes the video didn't have to have anything to do with the actual -- the content was important to get the user to stop and watch it as sort of the first step to the funnel. whether that was relevant to the actually acts why we're asking them to give us $35, didn't matter. and whatever video got the users eye balls is what we used. that is something that typical campaigns, i don't think we would be able to do. i think people would stop us and say well, that doesn't make sense, we're focused on that ultimate goal of getting the conversion and donation, so we're able to push through and do those things, additionally, because our ads got so many engagement, that led to even more reach with those ad units.

in swing states we would try to use persuasion style ads, message style ads, knowing that that ad unit is was going to be seen by our donor target and then their community. their friends. just by them engaging with that content. so we were able to garner incredible amount of kind of social impressions, more reach, with users that we weren't even targeting for donation. >> you double dipped. that's good. i'll come back to fund-raising, before we do that, i want to talk particularly to the cambridge guys about targeting. and you know, after 2012 and to some extent 2014, i think micro targeting was fetishized by some on the left there was intrigue about what cambridge was doing

with the psychographic profiling. i want to understand because we hear about that and we see what were very kind of generics, big theme, big emotion ads and you know daniel, i think it was daniel before saying, in some cases targeting, we need to actually stop being so efficient and we need to pull back a little bit. i want you to talk about how you used the tools that you have and just that field in general. >> yeah. i don't want to break your heart, we actually didn't do any psycho graphics with the trump campaign. i personally did woefully little on the digital side as well. most of my role was on the data side, rn side crushed it. we had tv ads budget from that. i meaninged the goat rodeo and zach went to new york. he was in my background,

periphery. so we didn't really use psychographics that much because we had to walk before we could run on this campaign. similar to the story that gary told, we started around the first week of june in san antonio with regulartively no data structure existing. we're parsing together csv files, it was a few weeks or a month before we could build a model and the emphasis was always on fund-raising. to gary's point on why he needed to raise money, every four, five, six, clips, it it was fueled by fund-raising, mitt romney had a fantastic dollar operation which could fuel the campaign. mr. trump's team did not have that luxury early on. every dollar that came in went to other parts of the campaign. brad was robbing peter to pay paul to get stuff done. on the targeting piece we're

talking about building a database, working with the rnc, the alamo base, and then leveraging cambridge's database, combining those three things together, building partisanship models, 12 issue sets, the basic building blocks you need from a campaign. over time we got into things like sentiment, basic nlp stuff, darren in the back. was working on a facebook spot late in the campaign that leveraged nlp technology. but we had five months to scale extremely fast and talking about doing siekographic profiles requires a much longer ramp time. >> the other thing that cambridge provided and the rnc provided me was if i want to make budget decisions every day which i was making lots of them, i wanted kind of two sets of data of what people thought was happening. i didn't want to make all my decisions off just one set of

data. at cambridge started helping us with polling internally, making what 1,500 live calls, 500 web -- it was 1,500 live calls per state, plus web and other stuff, i don't remember what the numbers was x we were bringing enough data in they could create models and the rnc provided, so i had two data sets to look at all the time. rnc say you are dun in georgia and cam rage saying no, i think we're up 3. i could say why is there a difference. i had a third set that were doing larger polls to find out what is going on on the ground. i was able to bring that data in and have a better picture of what i thought was happening. cambridge provided a full time employee that could sit next to me all day to put it in a visually sags, i could do it in a method so fast i could make a decision 15, 20 minutes because we were going so fast.

ability to digest that much data, even if sometimes it was the rnc data going back to cambridge and back to us, they were able to visualize it, so i could consume it to make better decisions. >> i think one benefit brad had he had control of research, data and digital and tech which is kind of a dream if you are someone who has been on presidentials before doing this. it gives you a ton of leverage and helps you make more holistic decisions where you are not fighting with the digital team to target the audiences you want to and you are not fighting with tech to build you a new page every day and spread operational control over those three or four wings, you can go to his .1500 surveys a week, across 17 battle ground states, mixed method -- live to sells, that goes into rerunning the models every wednesday, gets new audiences, gary gets new audiences and rnc is providing data and we're going back and testing the

digital performance off the back of that with google and facebook and also traditional survey work and dial testing. the campaign was consistently learning from itself which is very require and much more run like a business than most campaigns have been a part of. >> i might get gary to answer this, in terms of to borrow on the fund-raising. i think you raised a quarter of a billion, something in that range, which is phenomenal and i know there was days when you broke was it $7 million, some days, more than that, 10 maybe, nine? i want to ask two questions, one is a macro question. can you break down how you raised that money, i'm not going to get you to say we spent x with facebook, i know you are not going to do that, i want you to give me some idea of ratio between e-mail versus, you know,

ads bought versus x, give us some idea how you raised all that money. >> so the general break down was about 60 to 65% e-mail raise and then the rest basically advertising, some of that comes over the transom of people visiting the website obviously huge social presence. that outperformed my wildest expectations, there were a couple, one, two percentage points of the overall. e-mail i was happy to say, i was expecting to be 75%. i think gary and the cambridge team had an aggressive advertising style that really has never been seen before in politics that brought in so many extra donors initially and then e-mail was able to get them the second time. you know, the e-mail, like i was saying before, there are so many campaigns running in the summer of '15, we knew we had to have

an e-mail list, there is a time value to these assets, which means that the only way to build a giant list when you need one is either -- if you don't have one, is either spend a ton of money and catch up really fast but you are going to spend 10 or 20 x per e-mail, or build a time machine and go back and start the e-mail list from scratch, not scratch, the rnc had a good one when we came in, we built it up much larger. having assets ready was incredibly important. i think going forward in the next cycle, the next presidential, i think text is going to be more like 10% to 15%. >> talk about that a little bit. that's pretty interesting i think. >> sms, gary and the group ended up running a lot of that, but people were signing up, you guys had those -- the banners on this podium, all throughout the

primary, you walked in with like 600,000 text messages, i totally dismissed it and said there is no way we'll end up raising money on it, ended up doing several million i would say. >> towards the end it was a lot heavier and they had messages, mms messages, video messages where we raised half a million dollars, which was rather absurd. and you know, brad switched up the event, you had to start running your own event sign up where you are captioning mobile funds and that helps for growth in the msm list and also had a lot of protesters, they had to get verified through their mobile number. that was the goal. it also gave us a massive sms file which we raised a ton of money off of. >> combined together you have the e-mail list, the big tech message list. you can do more, that is why the

assets are so valuable, i would go to the political director and pns director and say the e-mail, the money that i need for e-mail, should you be arguing for too that will turn out voters persuade vetters. next to the candidates time and message i'm not sure what is more valuable asset besides e-mails and smss. >> you have to remember, too, the reason why they raised so much money is because donald trump was a really good candidate to raise money from. >> i'm glad you say that. the next question i was going to ask you is how much is represent my kabl and how much is because of donald trump. i want you guys -- >> and the three kids. >> yeah. you twisted me in, sir kels here. so if you are a campaign that is is maybe gearing up for 2018, what things can you take and learn from the trump campaign

and what things do you think were uniquely trumpian and cannot be so easily replicated. >> good candidates with good messages win election, what we do helps on the margins, my life before helping brad and cambridge was walker in wisconsin and parks who i think is somewhere out here doing my fund-raising for that there. i would have died to be able to say half the stuff that mr. trump would say in funds raising, a lot of typical republican candidates, others of you who worked for rubio, i see other folks in the room would have loved to say and have the kind of message and ten or mr. trump had in his fund-raising, that message matters and parks will tell you even subtle word choices mean the difference of 10% to 20% in e-mail on deliverability. mr. trump was the vehicle, but they were able to harness it. >> what was some of the messages

that resonated, what was some of the things that really popped. >> build a wall. >> give us insight to how that worked. >> one of our biggest days outside of the third debate was august 31st, the last day of the quarter, great day, also the day he went down to mexico, totally dominated the media. tv for us really felt kind of like, it would create an echo, whatever we were doing online would be juiced up if he is dominating the television news, all the users are watching, it's relevant and front in the face. 8/31, he was the president, he was playing that role. and he is also talking about an issue, you know, immigration, wall, that does really well for us. there is -- when i talked about earlier how we started using persuasion ads to raise money, you know, talking about immigration wasn't really our

persuasion ad, but we had an ad from early on, it was direct to camera, which we did a ton of these direct to cameras, i started going up to new york and writing scrips for the kids and mr. trump filming then. we had one eric trump was talking about the wall and it slaughtered for months, we keep running it and running it, most of our creative would burn out after a couple days because we were aggressive with pushing it. we would pull it down for a couple days and bring it back. we could be asking money for anything, that piece of content would fire the user up, ask them for a membership card, this or that, they're interested now. >> are we done? i will give you one thing that is interesting about the fund-raising, we had all this in one kind of shop. fund-raising, small dollars literally followed two days ahead of polling data. and that was really the truth. meaning you could take that

graph and show as fund-raising went up the polling data to media and things that got happened got them excited in the base and made them move to trump's column would vote with their wallet. >> when it plummeted after access hollywood did the -- >> as polling -- as the votes came, as you saw as we improved or the ebb and flows of the up and down of the campaign, fund-raising would match that. people vote with their wallet. it was nice with a small dollar fund-raising, you could see this. it was nice because i could see as the in goes as we saw other data showing the victory, fund-raising continued to grow as well and it went all the way to the last day and people were trying to vote with their wallets, when they weren't happy they would vote with their wallet. >> i want to ask you one last question. what is going to happen, brad, will you be working with kelly an. >> i will have a job outside the

white house. >> are you going to be the chief digital dude on the outside group? >> no. i haven't made -- right now i'm enjoying life, dinner with my friends across the street. so that's as much as you are going to get. i will have a job and it will probably be outside the white house. those are the ftwo answers to that. >> we're all fairly confident. thank you, guys, appreciate it. [ applause ] >> okay. we're going to show a few videos from the super packs for the last panel and then we're on to cocktails. >> why aren't i 50 points ahead you might ask. >> well despite insisting. >> i am a real person. >> hillary admits -- >> last time i actually drove a car myself was 1996. >> the clintons made $100 million.

>> we came out of thehouse not only dead broke but in debt. >> she is under fbi investigation, but -- >> people should and do trust me. >> so why aren't i 50 points ahead you might ask. >> 45 committees responsible for the content of this advertising. >> hi,heim ru i'm ruth, two cer things in life, death and taxes, one thing is certain for me, with lieu an bennett you get higher taxes. it rules out raising taxes saying there are times when you need the revenue, doesn't the government take enough already. meanwhile she is out there criticizing her owe po intent. whether it's higher taxes on families, higher taxes on businesses, you can couldn't on lieu an to lead the charge. virginia doesn't need lieu an bennett, learn for. >> my home, my family, and my

marriage, i have a lot of things to worry about besides some politician's future. empty promises of special treatment are the last thing i need. i want a leader who will fight for me and my family. from my home to my job, there's a lot depending on me. i know i can build a better life if we can finally get the economy on the right track and there is no limit to what i can do when my family, my job and our security are priorities for my elected leaders. that's why i'm supporting carla for congress. he knows i deserve equal pay for equal work and he understands when i succeed, my family succeeds and america succeeds. >> what is at stake in this election. it's not just who goes here, it's who rules here. the supreme court. the justice who guaranteed your right to own a gun is gone.

now, the next president's choice breaks the tie. four justices support your right to own a gun for self defense, four justices would take away your right. >> the second amendment is outdated. >> the right to possess a gun is clearly not a fundamental right. >> what does the second amendment mean to you? >> not the right of an individual to keep a gun next to his bed. >> and hillary says? >> when it comes to guns, we have just too many guns. >> the supreme court is wrong on the second amendment. >> hillary has made her choice, now you get to make yours. defend freedom, defeat hillary. the nra institute is responsible for the content of this advertisement. >> in iran, a woman's life is worth half her husband's, christians persecuted, it's illegal to be gay. people stoned beaten and hanged for what they believe, how they

were born and who they love. ted strickland supported giving billions to that regime. without concessions for human rights violations, when we had leverage, ted didn't stand up for the vulnerable. that's why we can't trust him to stand up for us. american unity pack is responsible for the content of this advertising. >> why aren't i -- >> okay. so brian we're going to start with -- can you hear me? okay. so future 45 didn't start running ads until later in the race at which point there had been hundreds of millions of dollars spent against hillary clinton. what made you guys decide to get involved at that point and how did you guys decide what kind of messaging to use, who you wanted to target and how you wanted to run your ads?

>>. >> hundreds of millions was probably an overstatement there. >> i was about to object to the premise of your question. i think at the point that future 45, we were formed in early 2015, but we certainly didn't spend the bulk of our resources until after labor day of 2016. at the time that we got involved, i think something like $150 million had been spent against trump by secretary clinton and her allies and i think maybe 8 or $9 million had been spent supporting mr. trump, mostly by the nra. so the playing field was wide open in terms of the types of issues you could talk about to define secretary clinton. from our point of view, we always start with research. so we undertook a very large research project, both on the subject matter and also on what voters cared about, then we tried to target our ads to the right audience. so we had ads as lee knows aimed

at millennials, ads that were aimed at folks who really cared about the women's issues, ads that were aimed at people who might engage with politics on a humerus basis, we did a lot of humorous ads and tried to be funny. like the supreme court and economy and so forth and tried to match those creative opportunities with the right audience. >> what was it like running ads for a super pack for a candidate who was already so well known in the media? was that part of the reason why you decided to mostly just go after clinton rather than airing ads promoting trump? >> we did both. the media liked to talk about the anti-clinton effort, but the reality is there were a lot of opportunities to positively talk about the president-elect and vice president elect. in fact, the ads aimed at economic opportunity and the

economic plan, i think were a real bulk of the advertising we did, especially in september and especially in pennsylvania, some of these will sound familiar, michigan, wisconsin, ohio, iowa. so we definitely focused on that. at a certain point, everybody understands the role of an outside group versus the campaign. at a certain point the campaign itself put up some really high quality positive ads aimed at talking about mr. trump's economic message. and at that point, when it became clear that they were carrying a very effective positive message, we then altared our strategy to go more negative. there was another group run by -- they had several really great quality ads as well. so we focused more on defining secretary clinton. >> how did you determine what

effect your messaging would have on down ballot races, they -- >> the map is the map. i believe that whoever won the presidency would also carry the senate. and one of the targeting opportunities was there were a number of voters who were really core trump voters who were not necessarily for the republican senate nominee. then there was another sub set that they were for the senate nominee but not necessarily for candidate trump. so we tried to aim our messaging both on air, but more significantly through digital means and then to a lesser extent mail and phones to marry up those audiences, if you could persuade a rubio voter to also vote for trump and vice versa, a trump voter to vote for rubio, we felt a rising tide lifts all boats, as todd used to like to say, hillary's slogan was

stronger together, you're view was we're better together. if we could be better together and get folks to turn out for the entire ticket, both the presidential candidate, senate candidate and my friend shields would say the house candidates. finally that was the reason why future 45 spent really the bulk of its resources buying national advertising. so we were up on i think it was 18 different cable and broadcast networks nationally, including $30 million in the last week alone as people were making their game day voting decisions. >> so just to open it up a little bit. this was obviously an unusual election year and one way was that we saw less spending on tv advertising on the presidential level, but more in congressional races. would you guys each talk a little bit about how you worked to make your ads kind of break through that noise, particularly on digital platforms. why don't we start with you, mike. >> with me, sorry.

we can talk about the ruth ads that's what we did, one of the things that we did. the concept for the ruth ad that you saw, really was me looking over the shoulder of my 14-year-old son who watches a lot of youtube content, and thinking through about how you have websites that are mobile sites that are different, and a lot of the content that we created in campaigns over the last two, four, six, eight years, we've really -- as we've -- the argument about whether or not we should put things online and what the budget should be, i think has changed, it's sort of been settled in campaigns to a certain extent that you are going to have to do this. now it is what does it look like in my opinion. and taking a television ad and cutting it into a 15 second and throwing it online was sort of 1.0. as we get to a different version of can we take the exact same research but do different creative that is the same message and the same -- to the same audience we are trying to hit on television, but do

content that works online, that is going to be in someone's feed that they're going to recognize, they're not going to swipe past because it's a political ad or it doesn't have clouds over the capitol with lightening, something you can't see, what is if that people look at when they're looking on small screens and what do they pay attention to. that's how we kind of came up with the idea of having ruth, who is sitting right here, talking straight to camera in a sort of familiar way and part of that too is we came to google and we said to google, look, you manage youtube stars and create these personalities, if we wanted to create a position at our -- we have a com director, press secretary, can we produce an online spokesperson talking to our audiences and trying to build a relationship with them the same way youtube stars do that and the personalities they build on there and so brian, our digital director, our

communications director, i have to say this, i sort of came up with that stuff and i was off doing other things and they actually executed all this stuff and made it happen. so -- they could tell you a little more, but they crushed if in terms of people paying attention, clicking, watching the whole video, the feedback that we got, how they moved numbers and so for us, you know, our approach was creative is king, buying online and all those sorts of things has reached a point where we can hire a lot of firms to do that very well, but what i'm not seeing as much of is people paying attention to the specific online creative and so that is what we focused on more at congressional leadership fund. >> i would like to jump in there, too. i think with what we did for the nra, we did two types of creative, i'm from tennessee, they say to make a good country song you need three cords and the truth, i think that works on digital creative. what we did was an authentic real person story with black screen behind them, nothing

else. something you can watch on the phone and -- if it wasn't that, the next thing we did was stuff that looked like a movie. because we're platform agnostic, most of our creative was designed to have parts that would work on all platforms and so we didn't want to interrupt people's entertainment each night as they watched whatever their favorite content was with advertising that looked like a piece of crap, piece of junk mail. we invested heavily, the ads that got $5 million put behind them, we had several for the nra, those looked like a movie, we spent the money to make it, we spent months in development and testing. one of the ad ds we made 12 versions of before we got it right, we went to focus groups in four states and kept getting it rong and trying to learn from our mistakes, you asked about the difference of how you decide what works online versus on tv,

we took a really blended approach, we think all these things work together. people use all devices, we don't pit flat form versus platform. we try to pervade with everything. we did 13 tractor trailer full of mail, it was a holistic approach. so i don't know that i would argue that you make those decision continue tians, you say how can this help me deliver that message. we thought youtube and google did a great job helping us own big events, we thought this was a race with two national stars and big events would matter. we were ready to go with youtube and searched traffic for conventions. we had specific creative that ran in the conventions, a violent crime survivor only ran during convention night when hillary gave her speech and we had a budget to run it. third debate i think we did probably took advantage for man anybody else did.

and captured i think quarter million, 257,000 searches on our topic when it came up during the debate. google and youtube helped us own it. when the world series came on, we bought ads but had a lot of search stuff going on. if you were searching for the team, candidates, second amendment, we knew that those big events were really important and we would have the whole country standing by some device. >> so i think that the question of how did we break through on digital, one of the things that drove us at american unity is we didn't really have a choice. i mean, my peers up here are running much larger organizations than american unity pack. i mean, we have our own budge, we have done $12 million as a super pack over the last couple cycles, you might do that in a single state. i think that for us, it's not only that we are a leaner organization, it's that we're single issue organization and so

american unity pack has a pretty focused mission of helping those republican candidates for congress who believe in lgtb freedom and so you know for us, we can also afford to play in those races, because we're dealing with a smaller universe. it was hard for us this cycle because we had so many allies, this is a good problem to have. but we had people like john mccain running in arizona, rob cordman in ohio, as you looked at the battleground senate contests a lot of the people running in those races were either incumbent, pro lgtb senators or a chaleninger we were excited about. looking at that at the outset, our team started to assess what could we afford to do, especially since we wanted to have a clear niche in helping these candidates. we knew we didn't have the budget to be, you know, overall a definer in the ohio senate race. but we knew that we could take on some part of the mission to

reelect rob portman and the clear path was digital. to refine that further, knowing if we were going to break through with our message, we needed to understand what was going to motivate the voters. so you are hearing themes, research, mull tiple version of ads. if you look at the light span of american unity pack even though we are an lgtb ad organization, at thispy about 93% of our dollars haven't mentioned lgtb issues, we're spending ads on corruption or attacking the democrats for being you lousy, this cycle we wanted to play with our issue a little bit and see if we could find an effective way to bring that in. we told ourselves we can't really do that unless we can prove that it's effective. then we can inject lgtb issues on behalf of republican

candidate in a general election and be successful in doing so. so we researched that. we worked with a group called the women's initiative, two ladies here, and we started to understand this group, the women's initiative is researching women's voting, motivations, opinions, it's a deep research program, you should learn about it, if you haven't and sub describe to it if you haven't already. we learned in that process, women voters, especially independent women voters were hawks, i'm dramatically simplifying this. we learned that they were hawks and there was a real foreign policy message. one of the ads you saw about the iran deal that is what we developed out of that deep research process was that we could talk about the iran deal, in a different way and instead of focusing on the nukes, focus on the human rights abuses of iran and the horrible things that the regime is doing to women and yes also to gay people and to show imagines of that and in the testing process, we

learned it just had amazing impacts with youngster left women, which is a prime group we want to peal away from the senate owe po nens, working with googt we were able to put serious money behind that in ohio in a first waive. do some brand lift survey advertising and understand where was it being effective. great recall was driving huge search traffic and again, we saw that that -- it was borne out by the real world advertising that our testing, you know, proved to be true, that young center left women responded to this image and baby boomers responded to the national security aspect of this message. we felt confident after running it in ohio we took it on the road to illinois and nevada, and i think there was maybe one other state, was there another state? pennsylvania on behalf of senator toomey. we couldn't have done that unless we had the deep research on the front end and the ability

to testing it live. we shifted resources and i think that was the great thing about working with google and also the nature of digital is you can make a lot of those judgment calls in real time. >> anything to add? >> i think part of your question was how did we know if what we were doing is working. and i think -- i think the moment where i thought, you know what, i bet trump thwins this thing, i think it was late october, but lee dunn, he sent out top five searches and i think trump was jobs, economic plan, sure there was a couple other crazy outliers on there, the one about secretary clinton is what happens if she gets indicted. how many foreign countries have given to her foundation. how many, blah, blah, blah, all five completely negative.

and what it showed was a lot of the messaging that all the outside groups were doing and especially the biggest megaphone, the president-elect was doing was breaking through to people and the trust issue really was being driven home. and so i think you got that real time feedback from the google platform. >> so brad, just to go back to you. the nra was one of the first groups that aired ads this cycle and became one of the top spenders. how did you decide to get in so early and what was the messaging that you found? >> a lot of that was due to the research from what went wrong in '12. we did an autopsy of our own efforts after the romney campaign loss and one of the things we learned is a lot of people we hoped to reach made their mind up very early and we didn't talk to them until very late. so we adjusted our schedule. we were the first group to run an ad in the general election, we did it before the republican

convention, that add got 2 million views on youtube and a million of those were purely organic. i think part of that is because it was con controversial and we managed the after us for shooting in a veteran cemetery and we baited it and took the hook and made the v.a. write us a letter and went on tv as much as possible. we are fine with managing chaos and controversy and experienced with doing it well. our post elect showed a 71% of voters decided before august. and so our creative was weighted pretty heavily to that. later as the campaign went on, one thing i would say that really worked for us with google and youtube because of their large scale in the rest of the economy beyond politics, trump coalition was very different. most of what we do on data as republican operatives is driven by a lot of things really grounded with our republican or independent voter history

targeting. but trump was going to win with a lot of people who didn't fit those heavy voter, heavy -- a lot of partisan data. we used google and youtube's resources of information from the other side to find people whose lifestyles matched. it was rural women. trump had a problem with women and 3% of the voters in the country who skipped the '12 elections were likely to vote this time and 79% of them backed the nra's viewpoint. most of them were rural women. if you were interested in do it yourself projects, camping, i'm not going to -- there are a lot of things that -- the list is long but we specifically aimed at that kind of data from google and youtube and you can only get it from a platform that big. >> great. to go back to you for a second, who is the target audience for the ads and how did you decide what content to focus on more than just the style itself? >> the target audience depended on the district. we had some audiences that were

built through the data trust. we actually worked with a deep root analytics and optimus and had on some of our districts we had them building out audiences for us. it really dependod which district we were going into and which audience we were trying to get to. the data shows the ruths ads crushed it with women which i don't think we would be shocked at that. and most places a big part of our audience and what we were trying to get to. that's a recurrent theme across all republican campaigns and target audiences. i'm sorry, what was the second question? >> i think that was it. >> oh, okay. >> thanks. and tyler, for you, you talked about sort of focusing on digital out of necessity in part. if you were to do tv ads with the content change or do you think you'd adapt it? >> it would have had to have changed. i don't know if stations would have run it. it almost didn't make it through

google approval. it was highly refined and our audience was -- and we were doing this probabilistically. it was overwhelmingly women, overwhelmingly independent and leaning left, soft partisanship scores. but we did build in to that a few other very interesting audiences. one of which we set out and lee and i talked about this very early on. we started doing this planning a year out. and one of the things we knew from previous ballot measure fights trying to win the freedom to marry in the states is that we knew that republican voters who had a preference for sushi were dramatically more likely to support gay rights. whatever that means. but we knew that. so we had a hunch if we built in a profiled audience and google has a custom sushi lovers audience, it's not huge, but it was highly efficient. you can't do that on television. that was on the iran deal

campaign. we had another one promoting equal pay for equal work for women. and we were really limited in that for starters we didn't have enough republicans who support the policy. if we had more republicans who support that, we could have done some of the most effective ads on their behalf. think about that for 2018. for this cycle, we weren't going to just air that to a huge audience. we know there are wrinkles on the right with a policy like that. but what we also knew from the research with the women's initiative is that that issue as a wedge is more likely to grab the attention of an independent woman than any other, you know, economic or pseudoeconomic issue that we're aware of. and so by being able to go into ohio again on behalf of senator portman or south florida on behalf of congressman cyrbello, we could target independent women audiences using digital that you can't do on broadcast tv or even cable. >> here's a closing question for you guys. there are any super pacs in the

democratic party that you thought did digital ads really well and really effective this year? >> i think they showed up today. >> okay. >> this is not quite answering that question, but some of the people at another conference have heard me bitch about this before. one thing that's frustrated me after this election and i give google all the credit for having the trump team here and people that have been involved in winning here so that -- and the press that's come here today to talk about this. but after the '12 campaign, you know, republicans had to face the music, even though brilliant things were done in the '12 campaign. the rnc data was good. a lot of good things happened but the democrats won so they're geniuses and we had to whip ourselves for years to get ourselves into shape again so that we could be ready for 2016. we did all that hard work with a lot of the work of the people in this room doing it. and we won. and so i'm waiting on the books

and the long articles and the magazine articles to talk about how smart we all were. and -- and the articles about how bad the democrats were because one of the frustrating things from the dccc and house majority pac are articles saying we were wrong about all these races but it's because our polling was bad. as if that's an excuse. we're held accountable for our polling. if the polling is bad it means our data back end was bad and the system created to build the polling infrastructure was bad. we got it right this time. our polling was better. voter scores at rnc were better. not only should we be celebrating that we got it right and more articles written about that. the democrats should be held accountable for how wrong they were and have an introspective argument amongst themselves about why they got it wrong. >> i was going to say, i agree with mike in that there's just to refine it slightly. the best thing the obama people did in '12 was the number of

creatives they did matching up with the number of audiences. we all read the articles. they were advertising on tv land aimed at folks who watch tv at 4:00 in the morning, reruns or whatever. republican candidates and republican groups and the party committees did a great job of huge numbers of creatives matched up with audiences. the one thing you asked about the democratic superpac, i wanted to answer it. i was stunned, and i don't know what the trump folks have said about this, but priorities usa basically ran the same playbook against trump that the republicans ran during the primary. ads and arguments and messages that didn't work during the primary, and they repeated it with $150 million of ads that didn't move the needle once. so i'm stunned that those donors to priorities usa haven't asked for a refund because it just really didn't work at all. >> you know, that's interesting. the ad that -- one ad they had had that worked really well was the -- they called it the role

models ad. a bunch of kids watching trump on television. i tested that thing everywhere. and it worked like a charm until hillary clinton came on in the last 15 seconds of the ad. and then it went in the tank completely. and i think they got that for a while. they cut it back to 30 and took her out of it. the last two weeks, what happens? they brought the version back with hillary back in it. i think that they couldn't get out of -- they couldn't out of the way of their own success in '08 and '12. i'm glad you had garrett and brad and garrett kobe up there. they built an e-mail fund-raising list bigger than hillary clinton's, and she's been working on it since watergate. it's -- and they did it in three months. and so i think there really does need to be a lot of coverage not only of the fact that a lot of guys on this side got it right but they ran 2008 playbook again. you can't do that, especially online. >> well, great.

thank you guys so much. [ applause ] that wraps it up. we're going to move to cocktails. we're going to play one last video that was the top forming youtube leaderboard video. if it doesn't make you want to drink and take a rest, i don't know what will, but it should be familiar to you. ♪ ♪ ♪ ♪ i've got some real estate here in my back ♪ ♪