Don't call it a survey

An email sequence aimed at chief executives was getting a new final step: one question about their own experience with the subject. Someone drafted it as a survey, and the room stopped. Not because surveys are bad, but because that particular audience deletes anything labelled as one. The word describes who the effort is for. A survey collects data for the sender. Feedback treats the recipient's opinion as worth having. Same message, same single question, different recipient.

The label is part of the offer

Wording debates in marketing usually deserve their reputation. This one does not, because the word here is not decoration. It tells the reader what role they are being cast in, and busy people read that role before they read anything else.

For an audience at the top of a company, the test is simple: does this ask treat my time as the input to someone else's spreadsheet, or as something being asked because my view carries weight? "Survey" answers that question the wrong way before the first sentence. "Feedback" answers it the right way. So does anything that signals a small, selected group rather than a mass send.

None of this makes a weak ask strong. It stops a strong ask from being deleted on the strength of one noun.

One click, not two decisions

The more interesting move sat under the question. The sequence needs newsletter consent from people who will never tick a box on a form, because ticking a box is a second decision and the second decision is where these recipients leave. So the enrolment rides along with the answer: the reader clicks the response they wanted to give anyway, and that same click carries the subscription.

The general rule is worth stating without the specifics. A consent step attached to an action the person already intends to perform converts far better than the same consent presented on its own, because you are asking once rather than twice.

And there is a hard line on it, which I would rather say than imply. The click has to state plainly what it subscribes to. A consent the recipient could not see is not consent, and the bill for it arrives later as complaints, as spam reports, and as a legal problem. Merging the step is legitimate. Hiding it is not, and the difference is whether the words next to the button would survive being read aloud.

Accepting the spill

A second decision from the same session belongs here because it shows the same instinct. The sequence reaches a share of people who are not in the target group at all, roles that look adjacent in a directory but buy nothing. The obvious reaction is to tighten the targeting.

The team decided against it. Tightening the filter also removes qualified people, and the loss of a genuine buyer costs far more than the waste of reaching someone irrelevant. Nobody tagged the off-target contacts separately either, because the tagging costs more effort than the wrong reach costs money.

That trade sounds lax and is usually right. Precision has a price, it is paid in the good leads the filter also catches, and most teams never calculate it. The discipline is to name the spill you are accepting, decide it is acceptable, and then stop treating it as a defect to be solved in every meeting.

What to take from it

• Read every ask from the recipient's seat and ask who the work appears to be for. Change the noun before you change the content.
• Never use the word survey with senior audiences. Feedback, or a named small group, does the same job without the reflex.
• Attach consent to an action the reader already wants to take, and make the words next to that action say exactly what it enrols them in.
• Decide what share of off-target reach you are willing to accept, out loud, and stop paying for filtering below that line.
• If the filtering would also exclude good leads, keep the spill. It is nearly always the cheaper error.