Polls as the key to victory: when to use vulnerability, benchmark and tracking polls

Know your opponent, know yourself, know your race, and you need not fear the result of any campaign for public office. This simple riff on Sun Tsp cliched "know thine enemy" strategy is the best way to prepare for and manage a political campaign. Polling is information that helps you run smarter. It should help your campaign choose to lead with your best strategies and tactics, and find the votes you need to win.

Budget

Usually the first question pollsters are asked is how much it will cost. My best advice is to think of your overall budget when deciding how much to polling to buy. You should spend between 10 and 15 percent of your campaign budget on polling. To know your opponent, you need a short vulnerability poll. To know yourself, commission a larger benchmark poll. And to know your race when it matters most, buy at least two tracking polls.

Vulnerability Poll

The most basic political research is a vulnerability poll, which answers the question, "Should I run against this opponent?" Schedule the poll as early as possible so the potential candidate has enough information to decide. You find out where your opponent is strongest, weakest, and if he or she is vulnerable to a generic candidate, your candidate, or another possible challenger. You will also get a preliminary reading on which issues might sway the election.

The vulnerability poll is short, less than five minutes. You want to see if the voters can identify you and how many can also identify your opponent. Poll each up-ballot candidate who might have coattails either positive or negative for favoribility. Test for the most important issue, action or personal characteristic most associated with your opponent in a positive and negative sense. Job approval of the incumbent and their re-election numbers are critical. Finally, test the head-to-head match-up.

Instruct your pollster to survey at least 400 registered voters. The data you get back will have a margin of error of just less than 5 percent, reasonable for making the decision to run. Spend about 10 percent of your polling budget. If your candidate does not have a reasonable shot, you have saved a lot of time and money.

Benchmark Poll

The longest and most expensive poll is the benchmark, based on answering the question, "What do they think about my candidate?" It can run 30 minutes. It is the most vital to running a successful campaign. While the vulnerability poll gave you a basic picture of your opponent, the benchmark provides more detail. This poll tests the most important aspects of the campaign plan: Message the me, targeting and media.

Benchmark polls not only report data on basic issue preferences and candidate support, they also highlight the intensity of those beliefs. Throw out selected opposition research bits and see if they stick. Depending on the comfort of your candidate, this might include anything from voting records to personal problems. Make sure that the research is factual, as you do not want the poll to become an issue. Just as important, make sure you ask about your candidate's assets and liabilities so you can find ways to exploit and manage each one.

This is the most important poll you will buy, so expect to spend about 50 percent of your polling budget. Try to get pricing that will allow for an extended survey of at least 600 registered voters for higher precision (margin of error plus or minus 4). The results will help guide the bulk of the campaign well into its closing days.

Tracking Polls

Inside a campaign, it is impossible to know for sure how you are doing with voters. Yes, there are signs such as fund raising, attendance at events and positive press. Without asking the voters themselves, you really do not know how to win your race. You might go a few months between the benchmark and tracking polls but at some point you need to know if any adjustments need to be made. Tracking polls answer the questions, "How am I doing, how is my opponent doing, and what do I need to do to win?"

I suggest at least two tracking polls, both of which should be shorter than the benchmark yet longer than the vulnerability poll. First, you should do a state-of-the-race test after a substantial effort at paid and free media. This way you can see if the media has been effective or if it needs to be changed. You want to test for knowledge and acceptance of campaign themes and detect any movement within key constituencies.

The final poll you should do is a pre-election ballot test about two to three weeks before the election. After you have made your adjustments to the campaign plan, this poll will help direct the final push to Election Day by refining final messages and direct where the candidate needs to maximize his or her time. Plan to spend the remaining 40 percent of the polling budget (or more if you exceed your fund-raising expectations) on tracking as many likely voters as possible.

Ultimately, how much polling you choose to buy has to do with your budget and changing circumstances within the campaign. In most cases, a benchmark is a "must-have," while some candidates with limited budgets will forego a vulnerability survey. Tracking is for the well-fed campaign team and usually one positioned to win. With a full program of polling you will know enough about your opponent, yourself and your race to win.

Michael D. Cohen, Ph.D, is vice president of Fabrizio, McLaughlin & Associates, a Republican research and consulting firm based in Alexandria, Va.

COPYRIGHT 2004 Campaigns & Elections, Inc.
COPYRIGHT 2004 Gale Group