How to Ask 'How Did You Hear About Us?' and Actually Use the Answers

03 September 2026·6 min read

Quick answer: Asking 'how did you hear about us?' is only useful if the answers land somewhere you'll actually look at again — a single spreadsheet is enough, no CRM required. Ask it casually, in your own words, right after someone books or buys, not as a formal survey. Then review it monthly, not just when you remember to, and let the pattern (not any single answer) tell you where to put your time and money next. 📈

Almost every small business asks new customers how they found them, and almost every small business does absolutely nothing with the answer. It gets a nod, maybe a mental note, and then evaporates. That's the real problem here — not that businesses aren't asking, it's that the answers go nowhere and get used for nothing. We started properly tracking this for our own leads a while back 💖, and it changed where we spend our time in ways gut feel never would have, simply because gut feel is built from whichever conversations you happen to remember, and memory is a terrible sample size.

What most businesses get wrong

The question itself is rarely the problem — most people do ask it, at least sometimes. The failure happens after the answer comes back.

No consistent place to put it. One answer lives in a notebook, another in someone's memory, another in a text thread — there's no single place to see the pattern, so no pattern ever emerges.
Asking it as a formal, awkward survey question. 'How did you hear about our business?' read off a script sounds like market research, not conversation, and people give vague, unhelpful answers to match the tone.
Treating every answer as equally reliable. 'Google' can mean a paid ad, an organic search, or someone typing your business name in because a mate mentioned you — without a follow-up question, all three get lumped together and the data means nothing.
Reviewing it never, or only once a year. By the time anyone looks at twelve months of scattered answers, nobody remembers the context behind any of them.

The tracking system (one spreadsheet, five columns):

Date | Name | Their answer (in their words) | Category | Booked/Bought? (Y/N)

The Category column is the important one — you fill it in yourself, translating their answer into one of a short, fixed list so it's actually comparable: Referral / Google Search / Social Media / Paid Ad / Local (drive-past, signage) / Returning Customer / Other. Keep the list short on purpose — more than eight categories and the data gets too messy to read at a glance.

The exact phrasing to ask (not a survey voice):
'Oh quick one before we go — how'd you end up finding us?'
or, in writing: 'Random question — how did you first come across us?'

Both sound like curiosity, not research, which gets you a real answer instead of a polite shrug.

The monthly review (10 minutes, same day each month):
1. Count entries per category
2. Compare this month's top category to last month's
3. Ask: does my current spend/effort match where the answers say people actually come from?
4. Note one thing to try, stop, or increase before the next review

How this plays out for different businesses

A referral-driven business that assumes most of its work comes from word of mouth: six months of tracking shows referrals are indeed the biggest category, but a fifth of 'referral' answers are actually people who saw a review online first and only asked a friend to confirm it was legit. That's a genuinely different insight — it means the reviews are doing real work too, and are worth actively asking for, not just relying on referrals alone.

A business testing a new price point on a core offer: tracking the source column alongside a 'booked/bought' column shows that leads from paid ads convert at the new price noticeably worse than referral leads do. That's a strong signal the price test needs to run separately by channel, rather than reading one blended number that hides what's actually happening underneath it.

A local service business coming out of a seasonal quiet patch: the monthly review shows local/drive-past enquiries spike hard in the two months before the busy season starts, well before bookings do. That's the exact early-warning signal worth watching for next year, so signage and local visibility spend can be timed to arrive a month ahead of the pattern, not react to it after it's already begun.

How to set this up without any special tools

Create the spreadsheet today, before anything else — a blank Google Sheet with those five column headers is genuinely enough to start. Don't wait for a CRM or a fancier system, because the habit of asking and logging matters far more than the sophistication of where it's stored, and a system you never open is worse than a plain spreadsheet you check every month.

Decide who asks the question and when — straight after booking confirmation, at checkout, or during the first proper conversation are all fine, just pick one consistent moment so it actually happens every time rather than only when someone remembers. Put a visible reminder there if you need to — a sticky note on the till, a line in your booking confirmation script, whatever keeps it front of mind.

Set a recurring date in your own calendar for the monthly review — same day each month, ten minutes, no exceptions. The system fails almost every time not at the asking stage but at the reviewing stage, because reviewing feels optional in a way that talking to a customer standing in front of you doesn't.

💡 Write down their answer in their actual words before you categorise it. 'Saw you on the drive to work' and 'Google Maps' might both get filed under Local, but the exact phrase tells you something a category never will — in this case, that your signage is doing more work than your listing is.

Mistakes to avoid

Don't skip logging an answer just because it seems obvious or unremarkable — the pattern only shows up with the boring, ordinary answers included, not just the interesting ones.
Don't change your category list every month — consistency over time is what makes the trend visible, and a shifting list resets your baseline every time you touch it.
Don't act on a single month's data — one unusual month is noise, three months in the same direction is a genuine signal worth acting on.
Don't ask the question then interrupt the answer to keep the conversation moving — the follow-up detail is usually where the actually useful part lives.

Frequently asked questions

Do I need a CRM to do this properly?

No — a spreadsheet handles this perfectly well for the vast majority of small businesses, and adding a CRM before the habit is established usually just adds friction that gets the whole thing abandoned within a month.

What if customers say 'I don't remember' or give a vague answer?

Log it as 'Unclear' rather than guessing or leaving it blank — an honest gap in the data is more useful than a made-up category, because it stops you drawing false confidence from a number that was never real to begin with.

How long before this data is actually useful?

Give it a genuine three months before drawing conclusions. Earlier than that and normal month-to-month noise can look like a meaningful trend when it isn't, and acting on it too soon can send your spend in the wrong direction.

Should I ask this of every single customer, or just new ones?

Focus on new customers — that's where the marketing insight actually lives. Asking returning customers is optional and mostly just adds volume without adding much new information, since you already know how most of them found you the first time.