Should You Trust Your How Did You Hear About Us Data?

A practice owner called us in the spring, pretty sure of one thing: his ads were a waste. "Nobody ever says they found us on Facebook," he told us. "When we ask how they heard about us, it is always a friend or Google." So he had cut the ad budget three months earlier. And now, oddly, the phone was ringing less, his Google traffic was softer, and the friend referrals he trusted so much had thinned out too. He could not figure out why.

We could. He had trusted the wrong number. The "how did you hear about us" answer is the most comforting data point in a practice, because a real human says it to your face, and it costs nothing to collect. It is also one of the least accurate. And when you make real money decisions on it, you can end up starving the very channels that were doing the work.

Why the answer feels solid but usually is not

Here is the thing nobody tells you: that question is a memory test, and human memory of a marketing journey is terrible. By the time a patient is standing at your desk, the path that brought them there might have started a week ago and touched five or six different things. You are asking them to compress all of that into one word, on the spot, while they are also filling out a clipboard and worrying about their appointment.

So they reach for the easiest answer. The marketing world has a name for this whole approach, self reported attribution, and it has been studied a lot. One widely cited analysis found that around 47% of people who picked the first option in a list were misattributed, simply because folks lean toward whatever choice sits at the top. Put "Google" first on your form and Google wins, whether or not it earned it. Researchers call it the primacy effect, and your front desk survey is full of it.

Patients credit the last step, not the first

The second problem is even bigger. People remember the human, personal moment and forget the rest. Picture a real path to your door:

• Tuesday, your Instagram ad shows up while she is scrolling. She does not click, but your name lands in her head.
• Thursday, a neighbor mentions she is looking for a new dermatologist. "Oh, I think I saw one nearby," she says.
• Friday, she googles your practice name, lands on your site, and skims your reviews.
• Sunday, she asks that neighbor if she has ever been to you.
• Monday, she calls and books.

Now the front desk asks how she heard about you. What does she say? "My neighbor." Clean, confident, wrong. The neighbor was step four of five. The ad that first planted your name gets zero credit, and the Google search that let her check you out gets none either. This is exactly why word of mouth almost always looks larger than it really is, and paid and social almost always look smaller. The channels that do the early, invisible work of making people comfortable with you never get the thank you at the counter.

And there are a lot of those early steps. Marketing research in 2025 pegged the average buyer journey at roughly 6 to 11 touchpoints before a decision, up from about 8.5 in 2021 as people bounce across ads, search, reviews, and social before they act. Healthcare is a high trust, careful choice, so patients often research even more. One survey answer cannot carry that whole story.

The channels patients never name are often the important ones

This is the part that burns practice owners. The stuff patients forget to mention is frequently the stuff working hardest. A 12 month study found that self reported surveys undercounted hard to track channels like social media, podcasts, and word of mouth by around 90% compared with software based measurement. Not by a little. By almost the whole thing.

That is what happened to our friend with the Facebook ads. His social media presence and ads were the reason so many people already knew his name when they searched or asked around. Kill that top of the journey, and a few weeks later the searches slow, the referrals thin, and the "Google" and "friend" numbers he trusted start sliding too, with no obvious cause. He cut the root and wondered why the branches died. If you have ever wondered why social media does not seem to get patients, this measurement gap is a big part of the answer: it is getting them, you just are not seeing it counted.

So should you stop asking? No.

Do not throw the question out. It still has a real job. For the channels you genuinely cannot track any other way, a referral from another patient, a sign someone saw, a talk you gave at the community center, the front desk answer is often the only signal you will ever get. The mistake is not asking. The mistake is treating that one answer as the whole truth and making budget cuts on it.

A few small fixes make the answer less misleading:

• Do not list "Google" first. Rotate the options, or leave it open ended, so the primacy effect stops handing free credit to whatever sits at the top.
• Ask it as a real question, not a checkbox. "What finally made you call us today?" gets a more honest answer than a list of boxes to tick.
• Let people name more than one thing. "I saw your ad and then my sister said she liked you" is the truth. A single choice forces them to delete half of it.
• Write down the words, not just the category. The specifics often reveal the real path better than the bucket you file it under.

What to put underneath the survey

The honest way to know where patients come from is to stop relying on memory and start capturing the source automatically, then use the survey to fill the gaps. This is not expensive or high tech. It is a handful of habits working together:

• Call tracking numbers. A different phone number on each channel tells you which ad, page, or listing actually produced the call, no memory required. We break down how this works in our guide to tracking where patients come from.
• Unique booking links. A distinct online booking or landing page link per channel means the system records the source the moment someone books.
• A source field on your forms. Capture the referrer automatically on your website forms instead of asking the patient to remember it.
• Platform data. Your Google Business Profile and ad accounts already report clicks, calls, and direction requests. That is real behavior, not recollection.
• The front desk survey, for everything else. Keep it for the true word of mouth and offline sources nothing else can see, and read it next to the hard data, not instead of it.

Put those together and the picture changes. Instead of one soft word, you get a real map of the journey, and you can finally tell the difference between a channel that gets the credit and a channel that does the work. That is the difference between guessing and knowing, and it is why we treat measurement as the backbone of any real patient acquisition system, not an afterthought. If you cannot see the whole path, you cannot prove what your marketing is actually doing, and you end up cutting on vibes.

The honest bottom line

The "how did you hear about us" answer is a hint, not a verdict. It over credits friends and Google, under credits ads and social, and misses most of the journey that led to the booking. Use it for the offline channels you cannot track, clean up how you ask it, and put real measurement underneath everything else. Do that and you stop making six figure decisions on a single word a patient half remembered at the counter.

Our friend turned his ads back on. Within about six weeks the searches picked up, the referrals came back, and the phone got busy again. Nothing magic happened. He just stopped trusting the wrong number and started measuring the whole thing.