WEBSITE CONVERSION · September 2026 · ~11 min read
The thank you page you are wasting
The thank you page is the one screen where a visitor is guaranteed to be paying attention. Most businesses use it to display the word "Thanks." Use it to say when you will reply, what happens next, and how to call you now if the job is urgent. It is also the cleanest place to fire conversion tracking.
On this page
- 01Why does the thank you page matter?
- 02What about the statistics saying most buyers will call a competitor?
- 03What should actually be on it?
- 04What response window should I actually publish?
- 05Why should the form redirect instead of showing an inline message?
- 06What should I never put on it?
- 07What to do this week
- 08When you do not need this
- 09Sources
- 10Related reading
- 11Questions about your thank you page?
Someone just gave you their contact details. That is the highest trust moment of the entire visit, and it is usually followed by a blank screen with a checkmark on it.
Meanwhile the buyer's anxiety starts immediately. Did that go through. When will they call. Should I contact someone else in the meantime. Every one of those questions has an answer, and none of them are on your page.
01Why does the thank you page matter?
Because the gap between inquiry and reply is where you lose deals you already won.
Local buyers rarely inquire with one business. They send three forms in ten minutes and go on with their day. The one who books the job is often just the one who felt most certain, and certainty is set on this page.
There is real research on the gap itself, and it is worth stating carefully because it is the most abused study in this category. In March 2011, James Oldroyd, Kristina McElheran and David Elkington published a short Vision Statement feature in Harvard Business Review reporting an audit of 2,241 US companies. Of those, 37% responded to a lead within an hour, 16% within one to 24 hours, 24% took more than a day, and 23% never responded at all. Among companies that responded within 30 days, the average response time was 42 hours.
The same piece cites a separate dataset of 1.25 million sales leads across 29 business to consumer and 13 business to business companies. Firms that tried to contact a lead within an hour were nearly seven times as likely to qualify it as those that tried an hour later, and more than 60 times as likely as those that waited 24 hours or longer.
Two disclosures have to travel with those numbers. Co-author David Elkington was founder and chief executive of InsideSales.com, whose platform produced the data. And Harvard Business Review is a magazine, not a peer reviewed journal, and this was a two page feature. Treat it as high quality vendor research published under an editorial masthead, which is what it is.
There is also observed behavior, not self reported, showing how little a late recovery is worth. Patient Prism's Dental Patient Access Report, covering 11,552,668 patient calls across 8,280 US dental locations in 2025, traces the funnel: of every 100 calls, 13 walked away without booking, 4.8 got a followup call, and 0.7 came back and booked. Fewer than one in a hundred who reached out and left ever walked back through the door. Vendor data, on its own customers, dental only. The direction is not in doubt.
Tell a buyer you reply within a stated window and you have changed what happens next. They stop shopping, or at least stop shopping as hard. Say nothing and they keep going down the list.
There is also a measurement reason. A thank you page is a distinct URL, so it is unambiguous proof that a form was completed. Inline confirmation messages are harder to track reliably and they break every time the form plugin updates.
A visitor on the thank you page has already said yes. Everything you put there costs nothing to deliver and reaches the most engaged audience you have.
02What about the statistics saying most buyers will call a competitor?
You will be shown them, and you should not use them.
The figures in circulation are specific and persuasive: that a large majority of consumers have abandoned a business after an unanswered contact, and that a similar majority say they will call a competitor if you do not respond. They are consistently attributed to a CallRail consumer survey of 1,000 US adults from September 2025.
We went looking for the primary document and could not find one. No CallRail hosted page stating those figures, no press release, no wire version. The problem is worse than a missing page: callrail.com returns an identical JavaScript shell with a 200 response for every path requested, including deliberately nonsensical control paths. A successful response from that host proves nothing.
So the figures are provisional at best, and they should not appear in a proposal or on a website until somebody produces the report.
There is a second reason to be careful with survey numbers of this shape, and it holds even when the survey is real. They measure what people say they would do. In the same set of figures, the share who say they would call a competitor is roughly four times the share who report actually having done so immediately. The gap between those two numbers is the honest discount rate on stated intent.
The defensible version of the argument needs no survey. If you reply late, some buyers will already have hired someone else. That is not a statistic. It is arithmetic about a person who called three businesses.
03What should actually be on it?
Six things, and it fits on one screen.
Confirmation in plain words, naming what they sent. "Your request for a kitchen estimate is in."
The response window, stated specifically. "We reply to every inquiry by the end of the next business day." Then honor it, because this is a promise and an unkept one is worse than silence.
An immediate route if it is urgent. A tappable phone number, right there, for the person whose water heater is currently leaking. Some inquiries should never have been a form.
What happens next, in steps. Who calls, what they will ask, whether a visit is needed, how a quote arrives. Reducing uncertainty is the whole job of this page.
Something useful while they wait. A short guide to what to prepare, your reviews, photos of similar work. This is where content actually gets read, because the reader is invested now.
And a way to correct a mistake. A phone number or email for anyone who mistyped their own number, which happens more often than anyone thinks.
If a visitor arrives here still unsure and starts hunting for answers, that is worth knowing before you add another widget to the site. Whether live chat helps a small business is a question the thank you page can partly answer for you, because it shows what people still needed after they had already committed.
04What response window should I actually publish?
Worked example
Measure your own before you promise anything. Six steps, one spreadsheet, an hour.
Step one. Pull your last 30 inquiries. Record the time each arrived and the time a human first replied. Not the autoresponder.
Step two. Take the median, not the average. One inquiry you answered a week late will drag an average past anything useful. Say your median comes out at six hours.
Step three. Bucket them. Suppose 8 were answered within the hour, 14 the same day, 6 later than that, and 2 never got a reply at all. That last count is the one that stings, and almost every business has one.
Step four. Add the outcome column. Of the 8 answered within the hour, say 5 booked. Of the 14 same day, 4 booked. Of the 6 late, none did. That is 63%derived against 29%derived against nothing.
Step five. Read it as direction, not proof. Eight inquiries is a tiny group and the difference between 5 out of 8 and 4 out of 14 could shift on two phone calls. You cannot claim a rate from this. You can see which end of the table your booked work sits at, and that is enough to act on.
Step six. Publish the window you already hit. If your median is six hours, promise the end of the next business day and beat it. Do not publish an hour because a study said an hour. A missed promise on this page is worse than a modest one kept.
Rerun it quarterly. After four quarters you have 120 inquiries and the buckets start to mean something.
Why should the form redirect instead of showing an inline message?
Three practical reasons, all of which save you trouble later.
Tracking gets simpler and more reliable. A page view on a unique URL is the sturdiest conversion event you can have. It does not depend on a script listener that a plugin update can silently break.
You get a place to send people. Ad platforms, email confirmations, and text messages can all point at that URL later.
And you get room. An inline message is a line of text squeezed above the form. A page is a page, and you can put the response window, the urgent phone number, and the next steps on it without cramming.
Keep the page out of search indexes so it does not turn up in results, and make sure it still says who you are, since some visitors land on it from a bookmarked link days later. Which server it lives on does not matter for this, though hosting decisions do affect other things worth getting right.
06What should I never put on it?
Do not put a second form on it. They just filled one in. Asking for more information here reads as a bait, and the details you want can come in your reply.
Do not put the full site navigation on it in a way that invites browsing away. A quiet header is fine. A wall of menu items sends people back into the site rather than into their day feeling handled.
Do not use it to sell something unrelated. A newsletter signup on a page where someone just asked you to fix their roof is noise.
And do not leave the default text your form tool shipped with. "Form submission received" is what a machine says. This is one of those places where the same discipline that applies above the fold on a first visit applies again: plain words, written for a person, doing one job.
07What to do this week
Submit your own form and look at what appears. If it is a line of grey text or a default message, you have found a fifteen minute improvement.
Run the six step response time audit before you write a single word of the page, so the window you publish is one you already hit.
Build a real page. Confirmation, response window, urgent phone number, next steps, one useful thing to read. Then set the form to redirect to it.
Confirm the conversion event fires on that page and appears in analytics. Do a real test submission and check, rather than assuming.
Then hold yourself to the response window you published. That sentence is now a commitment, and it is the part of this that actually wins jobs.
If you are running ads into this flow, check where the inquiries are coming from as well as how many, because geographic targeting mistakes can waste half a budget before the thank you page ever loads.
Be honest with yourself
When you do not need this
If your main conversion is a phone call rather than a form, this page barely matters. Put your effort into who answers and how fast.
If you already have a proper thank you page with a stated response window and tracking that fires, leave it alone. This is a one time build, not an ongoing project.
And if you are not actually meeting your response times, do not publish one. A promised window you miss is a worse outcome than an unstated one, and the fix is in the operation rather than on the website. That sequencing question comes up constantly, and what to fix first when conversion is bad everywhere is usually about the business behind the page.
Sources
- Oldroyd, McElheran and Elkington, "The Short Life of Online Sales Leads," Harvard Business Review 89(3), March 2011. Source of the 2,241 company audit, the response time distribution, the 42 hour average, and the seven times and 60 times qualification figures from 1.25 million leads. Conflict of interest disclosure: co-author David Elkington was founder and chief executive of InsideSales.com, whose platform generated the data. HBR is a magazine, not a peer reviewed journal, and this is a short Vision Statement feature.
- CallRail consumer survey, September 2025. Cited here only to say it could not be verified. No CallRail hosted page, press release or wire version stating the circulating figures was located, and callrail.com returns an identical JavaScript shell with a 200 response for every path including nonsense control paths. Do not use these numbers until a primary document exists.
- CallRail benchmark report, announced 14 January 2025, via Business Wire. This one is verifiable and is CallRail's own distributed statement, covering 1.1 million de-identified conversations. Included to show the contrast between a traceable vendor claim and an untraceable one.
- Patient Prism, "The Dental Patient Access Report," 2 July 2026. Source of the walked away and returned figures, from 11,552,668 patient calls across 8,280 dental locations in 2025. Vendor research on the vendor's own customer base, and dental specific.
- The response time audit is your own data. There is no published benchmark for local service reply windows, which is why the arithmetic above uses your inquiries rather than somebody else's.
Related reading
- Contact forms: every field costs you submissions. The screen immediately before this one, and where most of the loss actually happens.
- Writing a call to action people act on. The promise on the button and the promise on the thank you page have to be the same promise.
- Conversion tracking that survives a website change. The reason to redirect rather than show an inline message, explained properly.
- Measuring conversion honestly when your traffic is small. Why 30 inquiries gives you a direction rather than a rate, with the arithmetic.
Questions about your thank you page?
Email me at eric@seod.com with the URL of your thank you page, or tell me you do not have one, which is a perfectly common answer. I will send back the exact copy I would put on it for your business, including the response window sentence, ready to paste.
It takes me a few minutes and there is nothing attached. If your page is already doing the job, I will say so and leave it alone.
Or keep reading the conversion library.