Using Customer Surveys to Measure Offline Ad Impact

Some of the most valuable advertising data costs nothing: you just ask people. Well-designed surveys reveal what analytics can't see, especially for radio, TV and billboards. Here's how to measure advertising with surveys without collecting misleading answers.

This guide is part of our guide to measuring billboard, TV and radio advertising.

Survey type 1: "How did you hear about us?"

The simplest, most underused tool in marketing. Ask every new customer, at the same point in their journey.

Do it right

• Use a list, not only a blank field, so answers are consistent.
• Name channels specifically: "Radio (e.g., [station])," "Billboard or bus shelter," "TV," "Google search," "Instagram," "Friend or family," "Podcast," "News article or interview."
• Allow multiple answers. People usually encounter you several times.
• Include "Other (please tell us)."
• Make it required at a natural step (booking, checkout, first call) if it doesn't hurt conversion.

Read it with care

People tend to credit the most recent or memorable touchpoint. "Google" often really means "heard your radio ad, then searched you." If radio answers rise during a campaign, that's meaningful even if some customers also list search. Combine with branded search trends and tracked responses.

Survey type 2: Post-campaign customer survey

Send a short survey to recent customers or leads after a campaign.

Sample questions:

1. Before you contacted us, where had you seen or heard of [brand]? (select all)
2. Do you recall seeing or hearing an ad for [brand] recently? Where?
3. What do you remember about it? (open)
4. What made you decide to get in touch?
5. How likely are you to recommend us? (0–10)

Question 3 tells you whether your key message landed, which is valuable input for creative testing.

Survey type 3: Market awareness survey

To measure awareness among people who aren't customers yet, you need a panel survey of your target audience, ideally before and after the campaign, and in areas with and without it. That's a brand lift study; see brand lift studies explained.

Avoiding biased answers

• Don't lead: "Did you love our new radio ad?" pushes people.
• Ask unaided questions first, before mentioning specific ads.
• Keep wording consistent between survey waves.
• Watch who responds: your happiest customers are most likely to answer.
• Keep it short to reduce drop-off.

Privacy basics

Tell people why you're asking and how you'll use their answers, collect only what you need, and follow Canada's privacy and anti-spam rules when sending survey invitations by email. We'll cover the details in privacy laws for marketers.

Where to ask

• Online checkout or booking forms: a required dropdown
• Phone intake: a scripted question for staff
• In store: at the till, or a card with a QR code
• Email onboarding: a one-question survey in the welcome email
• CRM: a "lead source" field that staff must complete. See using a CRM

Wherever you ask, record answers in one place so you can compare months and campaigns.

Turn answers into decisions

Tally responses monthly by channel and compare them with spend. If radio mentions double during a campaign while blended CAC falls, you have strong evidence it's working. If a channel you spend heavily on barely gets mentioned, dig deeper with an incrementality test.