Increasing CATI response rates with simple wording changes (& not with incentives?!)

What makes a good phone survey introduction? It is important not to dismiss the significance of this piece of the phone survey puzzle. Advanis always ensures that we test various introductions to determine what works best for the particular study and audience. Indeed, we built our proprietary survey system to allow our consultants to do these tests very easily.

A solid introduction maximizes participation, increases engagement, and helps improve representativeness through higher response rates. Incentives are a common default to boosting response rates; however, incentives may not always be the answer.

On a recent business-to-business study, we tested two different introductions for a recruit-to-web survey.

• The first was formal, focusing on explaining the purpose (research topic; not a sales call), the time we needed today (5 minutes), and the draw incentive (15 prizes of $500), followed by a traditional question asking permission to continue.
• The second introduction was more informal, shorter, did not mention the incentive, and instead of length it mentioned that we had just 5 questions to ask today. This introduction also did not ask for permission to continue; instead, the interviewer would continue and end the call if the respondent asked us to.

The shorter, informal introduction with no mention of the incentive was the clear winner. This introduction yielded 4% more people who agreed to at least start the recruiting interview on the phone, asked for us to call back (i.e., made an appointment), or actually completed the recruiting interview and agreed to do the web survey. That is, intro #1 had a response rate of 25% and intro #2 had a response rate of 28%, a relative increase of 13%. Better yet, those who attempted to complete the recruit (answered the screening questions to determine if they qualified for the web survey) increased from 7% to 13%, a relative increase of 47% over the introduction that mentioned the incentive!

Because we had limited time in field, during this experimentation, many parts of the introduction were altered (rather than just varying one aspect at a time), which means that it is difficult to pull out the impact of each change separately; however, it is clear that experimenting with different introductions is worthwhile.

Alia Wright is a Consultant with Advanis. Since joining Advanis in 2008, Alia has been responsible for implementing and overseeing web, telephone, Intercept and IVR data collection along with implementing long-term tracker programs and online reporting capabilities.