N.C. Polling Firm Building a National Reputation

Posted February 13, 2009 at 6:02pm

A decade ago, when Dean Debnam’s wife was running for mayor of Raleigh, the successful North Carolina businessman was amazed to discover just how expensive it was to do survey research for a campaign.

He figured there might be a market for a company that could provide cheap polling data.

Two years later, Debnam started Public Policy Polling, a Democratic firm based in Raleigh that cuts costs on opinion polls by making use of automated telephone surveys, or interactive voice response polls, as they are referred to in the industry.

In a traditional survey, a polling firm develops and writes a questionnaire, which is then read to potential respondents by an employee of a phone bank. But by trading a human interviewer for a machine, Debnam’s company can charge $2,500 for a poll that might cost upward of $10,000 or even $20,000 elsewhere.

It’s a business strategy that is paying off. PPP spokesman Tom Jensen said that recently one North Carolina advocacy group that previously invested its yearly polling budget of $15,000 on one live poll, is planning on hiring the PPP to conduct seven polls over the next 12 months.

The use of IVR technology is still a controversial subject in the political polling world. Some traditional pollsters are quick to dismiss data provided by firms that use IVR and argue that automated polls aren’t very helpful when it comes to crafting or testing a campaign message. And despite the growing presence of automated polling firms in today’s political landscape (Roll Call commissioned dozens of polls through the automated polling firm SurveyUSA during the 2008 election cycle), the debate over live versus automated polling is likely to continue for some time.

These days, the PPP is also making a splash by giving away its product for free.

The company began pushing free polling information on its Web site in the leadup to the 2008 presidential primaries and decided to throw itself fully into the effort after many of the long-tenured traditional national polling firms proved to be so far off during the Democratic primary in New Hampshire.

“After everybody screwed up in New Hampshire, [Debnam] said, ‘Well, we can screw up as bad as they can. Let’s start polling nationally and be as prolific as anybody else,” Jensen said.

The PPP’s free polls didn’t stop with the end of the 2008 elections. Since the beginning of this year, the PPP is posting an average of three political polls a week on its Web site on subjects ranging from the stimulus package to the Colorado Senate race.

The company strives to make its polls timely, which helps in getting its data picked up by major news outlets such as the Washington Post and the New York Daily News.

Less than a week after Sen. Kit Bond (R-Mo.) announced his retirement in early January, the PPP put a poll on its Web site that showed Missouri Secretary of State Robin Carnahan (D) ahead of three possible Republican opponents in a hypothetical 2010 matchup. Five days later, the PPP was out with a poll assessing the Buckeye State Senate field in the wake of Sen. George Voinovich’s (R-Ohio) retirement announcement.

Jensen openly admits that releasing free polling information is another part of the company’s business strategy.

“It’s really helped us get our name out there,” he said, which in turn has helped bring in clients who are willing to pay for private polls that don’t get released. “We have probably multiplied 10 to 20 fold the amount of private polling we’re selling out of state since we’ve started doing this.”

It has also helped that the PPP’s public polling has gotten good reviews for accuracy in well-respected publications such as the Wall Street Journal.

Jensen said that last year the PPP had more than 50 Democratic clients and did private polling in 36 of the 50 states.

In an interview last week, University of Michigan political science professor Michael Traugott pointed out that the PPP’s business model is not new.

“This model of distributing poll results for public consumption in order to build your private business goes all the way back to the beginning of the industry,” Traugott said. “You can think about the Gallup Poll and its relationship with the Washington Post in the 1930s. … Where firms poll on issues of the day, release their results, try to build brand recognition and then to round up private clients, which is where they make their money.”

Traugott, who once coined the acronym CRAP — computerized response audience polls — to describe his belief that the methodology used by IVR companies is suspect, is still wary of automated polling. A main reason for concern, he said, is that while IVR polling companies work hard to merge voter registration information with phone numbers, a computer can’t discern who may pick up the phone when a polling call is placed.

“What does it mean to have a number for a specific person but to call them and not know who is on the other end of the phone?” he said.

Meanwhile, some more traditional pollsters point out that IVR polls and live polls provide different services for political campaigns.

Jensen acknowledged that IVR companies can’t do very detailed issue polling, and the PPP tries to keep its polls to 20 questions at most.

Democratic pollster Fred Yang of the Garin-Hart-Yang Research Group said, “There’s a lot of other things you need to know from polls that because of the limits of [IVR] technology you can’t get into a 20-question survey.”

Yang, who believes that the PPP has both a good product and a good business model, pointed out that firms like Garin-Hart-Yang “not only strive to give accurate trial heat numbers but also give our campaigns and clients the sense of strategy, message and tactics.”

“We can’t do your 100-question benchmark polls, by any means,” Jensen said. “But we helped a whole heck of a lot of folks in the last month of the [2008] campaign figure out, ‘Do I need to put $50,000 of my own money to put myself over the top? Or do I need to stop spending my own money because I have no chance? … We were telling them for 2,500 bucks where they stood instead of them having to pay somebody $20,000 to find out where they stood.”