Understand Public Opinion


Traditional polling is getting more expensive and yielding less accurate predictions.1 There are alternative means of gauging public opinion emerging, whether they rely on online surveys (YouGov, Survey Monkey, etc.) or social media activity (Zignal Labs, Kanjoya, etc). Each of these mechanisms has challenges – unrepresentative sample sets that often put more weight on the loudest voices, incapacity to get responses to precise questions, treating what is not being said as information, etc. – but they are getting more sophisticated quickly. For the most part, such tools have been adopted more readily in the commercial sector and providers are likely to make the bulk of their revenue there, so sacrificing some revenue in the political arena could be possible for these entities if other benefits outweighed the costs.

The data generated by polls, however, is only part of the value pollsters provide. Pollsters are often critical to developing and testing which messaging resonates with prospective supporters. Determining what questions to ask and how to interpret the results is likely to remain a highly sought after and highly compensated skill.

At the same time, there is growing public wariness of and frustration with candidates who seem to lack genuine conviction or present messaging that appears overly reactive to polls.


  • Could the alternative mechanisms for understanding public opinion develop cheaper ways of achieving comparable results, so that campaigns with less financial resources could compete? To accentuate that effect, could widespread low-level engagement2 give alternative service providers incentives to offer low-cost campaign versions of their tools by bringing new users/subjects to their services? (Increase the ROI of People Powered Tactics)
  • Could watchdog groups be cultivated that mobilize widespread low-level engagement by constituents to effectively and responsibly incur a cost to candidates who rely heavily on expensive pollsters to shape their messages? (Decrease the ROI of Conventional Tactics)
  • Could systematic research comparing these new methods with traditional methods in other contexts show comparable results at lower costs and influence campaign practices, particularly amongst challengers? Could an industry “seal of approval” be granted to pollsters that comply with certain ethical standards, including pro bono work for less well-funded campaigns? (Influence Campaign Culture)

What the Domain Experts Say


Kevin Collins

Director of Research, Analyst Institute
In short order, pollsters will need to figure out how to make the transition to online interviewing, before falling phone response rates render phone surveys unreliable. We are already seeing a dramatic growth in online surveys, from Pew’s American Trends Panel experiments, to Survey Monkey’s entry into election polling, to the rise of new online-only polling efforts like Morning Consult and Civiqs (and even Microsoft through its Xbox platform), all joining the ranks of established online polling operations such as YouGov and GfK’s KnowledgePanel. However, while all of these surveys use similar means to interview the electorate, their methods for sampling and recruiting respondents vary widely. This diversification of methodology creates new opportunities for disruptive innovation as campaigners find ways to advance online research methods that already are useful for national-level surveying and in the process making them useful for congressional-district-level surveying.


Mindy Finn

President and Founder of Misschief Media / Former Director of Digital Strategy, Mitt Romney for President
With technology, we may understand the hearts and minds of more voters, faster, cheaper and with greater accuracy. When campaigns see the value of a continuous, real-time read on the electorate, versus a static poll, they should develop a new habit of listening to the voice of rank and file voters.


Kyle Layman

Executive Director, Independent Lines Advocacy / Former Chief of Staff for Congressman Raul Ruiz
It’s no longer a question of whether we can improve public opinion polling; it’s an issue of how fast our ability to innovate will catch up with a rapidly changing world. Recent failures to accurately predict the outcomes of elections, for example the KY Gubernatorial of 2015, highlight the vast inadequacy of traditional methods of research. The change is necessary now, not just desirable, and with it will come real-time feedback and greater influence of the public in shaping the landscape of modern campaigns.


Jim Pugh

CEO, ShareProgress
I believe we’re extremely close to having online public opinion measuring that works as well or better than traditional polling, for a tiny fraction of the price. The major cost here will be in crafting the surveys and analyzing responses, although much of that process can likely be automated for standard opinion polls.


Kahlil Byrd

CEO, FPPCO LLC (Forward Progress in Politics) / Former CEO of Americans Elect
We are experiencing a data revolution in politics. Except now, big presidential campaigns aren’t the one’s driving the change. It’s an emerging community of data scientists who understand citizens more than candidates, operatives, and – yes – pollsters.


Hahrie Han

Anton Vonk Associate Professor of Political Science, Univ. of California, Santa Barbara
Any one person is a complex mix of shifting preferences and identities that our polling has, in the past, struggled to capture. We are just beginning to learn the dynamic ways that democratic (small d) organizations can both influence and understand the complex preferences and identities that people bring to the political table, making this an area ripe for new thinking and innovation.


Ethan Roeder

Former Director of Data, Obama Campaign / Executive Director, New Organizing Institute
“Donald Trump is a creature of the polls. He is his numbers. But he is only a sign of the times. Turning the press into pollsters has made American political culture Trumpian: frantic, volatile, shortsighted, sales-driven, and anti-democratic.” (From The New Yorker)

Polls are bullshit. Even when scientifically accurate, they are nothing more than the real-time manifestation of a history written by the victors. If you want to disrupt the public opinion space, fund technology that intentionally skews poll results to intentionally tell more loudly the tale of those who are marginalized, those who are oppressed, those who are without a voice. Accuracy in polling is only a sharper blade with which to separate individuals from the complexity of their realities.


Megan Garcia

Senior Fellow & Director, New America CA
With the rise of mobile technology, traditional polling will be challenged by apps that gather in-the-moment opinions from a variety of voters. This technology can be shaped to provide more information to a greater number of candidates, or it can be used to entrench those who already hold power. It’s important that the technology be used to (1) gather more data from more Americans, unlike traditional inaccurate polling; and (2) to broaden access to data about what policies Americans want.


Costas Panagopouos

Visiting Professor of Political Science, Yale University
Politicians can only be responsive to the public if they know what the public wants. An accurate portrait of public opinion is essential as a benchmark against which to measure responsiveness. Such a portrait of public opinion would reflect the views of all citizens writ large, rather than just those who are most engaged.


John Fallone

Executive Director, Lincoln Network
Understanding public opinion is the most important facet of the People Powered Playbook, in the sense that if the republic is to be more representative, then candidates NEED to understand public opinion. Additionally, candidates need to be incentivized by public opinion to adopt a position that is representative of their constituency. To accomplish this symbiotic relationship, inclusion of the entire voting population in elections is necessary to ensure positions of the general public are those that matter, and not the positions of a small subset of likely voters.


Mark Hanis

Consultant to Clinton Foundation and Charles and Lynn Schusterman Family Foundation
The shift to a more dominant mobile platform-led space will provide new opportunities to make campaigns more people powered (see Brigade Media as an early adopter of this).

Expert Summary Analysis

All answers: 1 = strongly disagree / 5 = strongly agree

How much do the conventional tactics for achieving this objective influence how reliant campaigns feel on narrow subsets as opposed to broad cross-sections of the electorate?

Average: 3.280%

How much do you think the tactics for achieving this objective are likely to change in the coming decade?

Average: 4.280%

How much opportunity do you see for advancing more People Powered tactics for achieving this objective via each lever?

Increase People Powered ROI

Average: 3.280%

Decrease Conventional ROI

Average: 2.330%

Culture Change

Average: 3.280%
  1. Read: “Polling Issues Getting Harder But It’s a Vital Check on Power” / fivethirtyeight.com
    Read: “Assessing the Representativeness of Public Opinion Surveys” / people-press.org
  2. We believe finding new tactics for achieving campaign objectives that rely on low-level investments of time or money from a broad diversity of citizens are central to increasing the viability of a competitive People Powered Playbook. If the conventional approaches have relied on narrow subsets of the electorate (wealthy donors, elite service providers or other potential gatekeepers) that change would be all the more significant. You will see this term “widespread low-level engagement” used throughout the analysis to reflect that aim.