Let’s assume that you have conducted a survey previously and are about to embark on the second (or third) in the series. It’s worthwhile to ponder for a moment the order in which you email invitations for the latest iteration.

Of course, at the relaunch of the survey you can fire off all your email contacts at once. Many sponsors shoot the works and use their bulk email software to write everyone and their brother, especially if they want to get the report out quickly. And then they usually hit up the same group a few times more at intervals. If they are more fastidious, they might drop from the reminder pings those who submitted responses or opted out.

But a more measured series of email salvos may result in a higher number of responses. Here’s why.

A step-by-step schedule of partial mailings allows you to build your respondent numbers and facts, so that when you invite the next tranche, presumably the harder-to-get and persuade invitees, you have a more persuasive story to tell. Each subsequent email blast supports a more fulsome and persuasive set of respondents as you have accumulated more and can describe them by demographic statistics. You might do more work and link to additional information on a blog about the growing cohort of participants. You could break out the participants to date by size, industry, position or other demographic. Doing so cuts two ways, however; some invitees might skip the survey because at that point they see the details it has too few participants in their desired demographic (“Hmph! Only 30 firms larger than 350 lawyers”; others might be motivated by the turnout (“20 from Texas; I’m in!”).

The best way to kick off a survey is to go first to those who took part before. You slipped by their spam filters, established some credibility, and might even have had a short email exchange. They know the time it took to complete the questionnaire and the value of the report. The downside is that they may decide that little has changed so why bother to step into the ring the second year. Or, shudder, they might have decided the survey report doesn’t justify the time would spend filling out the survey.

But even if you start with your incumbents, the first email to go out might aim at smaller cities or companies with lower revenue or lower level employes – build up to the bigger fish. Or you might tailor the subset emailings to metro areas. Whatever the segmentation of your email invites, raw numbers of respondents can impress some who are on the fence. If you are a general counsel or managing partner, would you rather invest your time on a survey with 22 respondents or one with 122 respondents?

Once you have exhausted your compilation of email addresses, you might turn to vendors and others who have an interest in the data and can supplement your efforts. They can reach another set of possible participants.

I’ve also found that a first wave of email invitations and responses may uncover a typo in the questionnaire, an unartfully worded question, a missing multiple choice (because too many chose “Other”), or a needed instruction. You can improve your offering for the second and subsequent waves.

If you send out emails to separate, randomly chosen lists of invitees, you can also try a form of A/B testing. Vary an aspect of the message and look back later to see which version elicited more responses.

Here’s the fundamental point of this technique: the more preparatory steps you take, the more the later invitees will respect your effort and feel inclined to leap on the dogpile. However, if you bulk up later email invitations, a tension arises: the longer the email message, the less likely the recipient will wade through it. So, if you add details about the number of respondents so far and statistics on their size, position, industry or whatever, you stretch out the email text. Now, you might create trenchant bullet points, but the effect is that the value of the additional information in persuading invitees to take part is diminished if they are deterred from reading by the volume of all the information. TLDR – shorter is sweeter. This topic also resonates with the strategic choices of when to send reminder emails.