How to keep polling costs within your overall budget

There is never enough money. Even the best-funded campaigns are always asking donors for more money. Both major candidates for president will spend more than $200 million this year, and each will ask for money until the end of the race.

In planning your overall campaign budget you will always find ways to spend money on events, signs and advertising.

Since polling costs can vary widely, it is important to examine how to keep polling from taking over the budget yet spending enough to get the information needed to win.

View Polling as Percent of Overall Budget

The first rule of thumb is to set the polling budget as a percentage of the campaign costs. Pollsters usually charge by the survey. So when you are looking to hire a firm, you should first decide how much you expect to spend. Generally, most campaigns should spend between 10 and 15 percent of their budget on polling. Therefore, the $1 million campaign should spend between $100,000 and $150,000 on polling. Set the percentage that makes best sense for your campaign before you interview pollsters so they have a fair sense of what they can offer to fit your budget.

Get All Costs Up Front

The second-most important decision is to hire pollsters who will be up front with all possible costs. Campaign committees are helpful in directing you toward preferred pollsters. But also open the search to other firms. Balance the suggestions of party officials with how much of a priority your race will be to the firm.

Demand all costs up front and make sure that everything is included. Some of these costs could include voter lists, multiple presentations and pollster travel. You might want to see a specific subset of voters that could make the difference in your race, and you do not want to pay extra for different cuts of the data. You do not want to be surprised by costs for things you thought were included in the price.

Select the Right Polls

Next, you need to decide what you need to buy. Are you in the race no matter what and strapped for cash on-hand at the beginning of the campaign? Ditch the opponent vulnerability poll and redirect funds to the baseline poll, which will give you most of what you need through the majority of the campaign. [See Camapigns & Elections, July 2004, p. 35] Do you expect a close race or is the party registration strongly in your favor? If you are expecting to win big then consider saving resources for mobilization efforts. If it is going to be a targeted race, you need to save more for tracking late in the campaign so adjustments can be made in time to matter.

Limit The Number of Questions

Pollsters have a financial interest in you asking more questions than you need. This goes well with your interest in wanting to know as much as possible. Put them together, and you have a potential budget issue. The best pollsters will help you keep a lid on the costs by steering you toward questions that will really matter to your campaign. Remember, if it does not help you design a fund-raising piece or an ad, or position your candidate well in the debate, drop the question.

Select the Right Number and Mix of Interviews

The final piece that can pry open an unexpected hole in your campaign budget is buying a poll with more voters than you need. Again, pollsters have a financial stake in the number of interviews completed, and this is the easiest way to blow up the budget. Adding only 50 more interviews can add hundreds of dollars to the cost of the poll, so you have to be careful. The important thing to understand is that the more interviews completed, the better the precision of the poll. This is the margin of error that is always cited in media polls.

Here is a list of common margins of error based on the number of people interviewed.

* 300 = 5.7 percent (use with a vulnerability poll--only to decide whether or not to run)

* 400 = 4.9 percent (the low end of what you should do for strategic purposes--not for publication)

* 600 = 4 percent (the low end of what is acceptable for the press)

* 800 = 3.5 percent (acceptable by the press; good internally if you have several groups to compare)

* 1000 = 3.1 percent (when released to the press, the numbers are usually accepted as bullet proof)

The final choice you have to make when watching the overall budget is what type of voters should be interviewed. Registered voters always are cheaper to interview than likely voters because there are more of them. For vulnerability polls, survey only registered voters, because it is usually too far away from Election Day for people to accurately predict how likely they are to vote. As the campaign gets closer to the big day, interview likely voters, as they are the ones who will make the difference.

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

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