To submit requests for assistance, or provide feedback regarding accessibility, please contact


Polling and Focus Groups

David Axelrod and Karl Rove

Lesson time 20:03 min

David and Karl trade off teaching the importance of polling and focus groups, while also highlighting the limitations of these types of research.

David Axelrod and Karl Rove
Teach Campaign Strategy and Messaging
Renowned presidential campaign strategists David Axelrod and Karl Rove reveal what goes into effective political strategy and messaging.


There are several different kinds of polls. There's what we call a baseline poll. It's generally a pretty long instrument done at the beginning of a campaign to get a baseline of where things are. During the campaign, you may update that. We call the updates brushfires, where you go in and do a shorter instrument. What are your favorable and unfavorables? And what's the ballot question? But may also include new questions that have emerged about an issue of the moment. But they're typically smaller in size than the baseline poll, but they have a number of questions on them. Then we have tracking, which is generally done at the end of the primary or the end of the general election. It has a smaller number of questions. The samples tend to be a little bit smaller. They're spread over several days. So that you might ask, say, 150 people a night, and you have a rolling average where you take, say, three nights of those people, and then you do a fourth night and drop off the first night. But what you're looking for there is movement. What's the ballot looking like? What's the information flow like? Are they hearing more about you? Are they feeling better about you? Are they feeling better about voting for you? And what are some of the issues that are popping up here at the end? But relatively short and fielded during a relatively short period of time. These are all things that we call quantitative polls. They involve numbers. We're asking a random sample of people, adjusted so it matches the demography of the district and the demography of the likely turnout, and we're trying to get a sense-- numbers and so forth-- of how people feel about you and your candidate and your issues. We also have what are called qualitative instruments and of principle one of these is a focus group. This is where you get a group of people-- very hard to do, so you generally have to use somebody who is an expert in this who has this as their career or their company's focus-- you get a group of people who represent some common view, like undecided voters in a particular area or soft Republicans or soft Democrats who like the candidate of the opposition party. But you get a relatively what you think are one of the key outgroups that you've got to understand better and deeper than the rest of the electorate in order to figure out how to them. And you get them in a room, and using a professional moderator, you walk them through a discussion about the campaign and the candidates and maybe even the advertising. [LIGHT MUSIC] - One of the mistakes that people sometimes make in campaigns is to assume that what worked in some other race or at some other time will work now. And so research becomes very important, and by research here, I mean opinion research-- polling; if you can afford them, focus groups-- to really test the array of messages and get a clear sense of where you're starting. You want to benchmark poll. Even if you're running a small campaign...

What it takes to win elections

David Axelrod and Karl Rove reach across the aisle to offer an inside look at winning campaign strategies. The respective architects of Barack Obama’s and George W. Bush’s historic election victories teach how to develop a campaign platform and reach an audience with consistent messaging. Find the inspiration and tools to get involved at any level, or simply become a more informed, engaged citizen.


Students give MasterClass an average rating of 4.7 out of 5 stars.

This is two people who know their profession. One works for good and the other for not so good.

Great Masterclass. I learned a lot. I learned things that I can apply in my line of work.

Always loved politics and this master class gave me unique insights into how it works. The passion for their work is obvious and very inspiring. I also see many things in political marketing and communications that I can use in my day job as founder and CEO.

It was extremely interesting to hear about campaign strategy from two very experienced practitioners. Thank you to David and Karl!


Jenny C.

Karl Rove is a much better speaker and communicator than David Axelrod. Axelrod says "uhh" every several words - it's a terrible habit for someone with his credentials.

Terri R.

So far... so good. I have a strong desire to run in the upcoming election. I believe I have what it takes. I never really voted in most primary elections so that is going to hurt me. I have voted in every presidential race. I have been studying US History/Leadership and back in the day the party was called Democrat/Republican. I Identify with both parties in some ways but I understand I would have to make a choice. I would choose independent. I believe my best shot to win in the upcoming election would be to run against both Biden and Trump. They are both dirty old men and I am a young and clean woman. I believe this could be the perfect storm for a candidate like me to come in and not only win but win by a landslide. I am running for the my childrens future and their childrens future. Would you guys vote for me? Any help you can give me would be greatly appreciated. I believe our lives depend on it. I wrote a book called, A True Unicorn... In the book I describe myself as, "The mother of Dragons, and breaker of Chains." My children are the dragons and the chains are the current government officials who are corrupt by greed, ego and the ugly desire for power to control people.

Martie A.

So what I wonder is how people are picked to do Ethno-Diaries. That sounds like something useful to participate in for an election.


It seems that all of his manipulation is designed not to get your message out, and the better-liked message wins. Rather, tell any group what they want to hear, even if the candidate doesn’t plan on following through. Rove is a master at manipulation, but clearly with zero integrity. Smear someone to move Yucca mountain fear? No problem. Bleh.

A fellow student

I wonder if personal blogs could be a source of data. I'm not sure how many personal blogs are out there but it might be interesting to compare them.

A fellow student

The main criticism that I would have here is that you can’t have the tail lead the dog. If you are a political leader you have to do what you believe is right to solve the particular problems that your country faced at any given time. Most people are not farsighted enough at any time so see that, and besides, you are the leader it is your job to show the way. Doing what polls good may win elections but it doesn’t solve problems. Karl and Dave, great tactics, but you are going to have to leave the vision part to leaders who want to solve problems. The Clintons were a good example of doing what polled right but the vision was absent.

Ron H.

When I ran, I had nearly nothing of a campaign budget and made serious mistakes about where I allocated my meager resources. I could not afford to run polls, or focus groups. I still conducted my own type of polling early in the Primary race and that was just by going out and knocking on doors and listening as much as I could given my ability to talk, talk, talk. I also knew that nobody whatsoever knew who I was other than my immediate neighbors within the District. So I was pretty sure that had a professional polling company conducted one, the answer to one of the questions would be "Ron who?" It was because of that, that I sought out every possible event, meet and greet, large public event and just got out there. Even the Party didn't take me very seriously, until the day after the Primary when I had won. I do think this is one of the best lessons so far however, especially the cautions about the data you gain from polls and focus groups.

Tony S.

The insight on the ethno-graphs technique was really fascinating. It seems like a great way to balance your ongoing assumptions with your constituents. In my experiences over the last election cycle (granted my first really watching for this stuff) I saw many campaigns running pretty blindly with their own assumptions. It reminded me of a company designing a product from the engineers perspective as opposed to the consumers. I also liked Mr. Rove's observation that too many campaigns over-empower focus groups with authority to drive campaign messaging when the information gathering tool itself can be so unstable. It does not seem like a very good tie-breaker tool. Insetead, it just adds color to the arguments and position that are developing which should likely be used to re-center the consistant decision making group and take a re-vote. I am now starting to clearly see Rove's central theme of developing a quantifiable rubric that can validate or invalidate your assumptions in planning throughout the entire campaign and a central (well understood and followed) process/mechanism to respond that is consistant. Great segment for me.

Bob S.

Wow, the power of this lesson was hearing how great focus groups are from Axelrod and then getting punched in the gut by Rove lol. Of course the devil is in the details. Focus groups are good but make sure they're stocked with more informed people who are motivated by the content not the $50. Have a good moderator to keep the loudmouth at bay. Oh! Don't forget fresh coffee and donuts!

Isaiah P.

I agree with Rove on this one. Polling works better. Trump lost the popular vote. Most people thought it was impossible for him to pull off rust belt so the frequency of polls in those areas were scare towards Nov 8th