How fast someone replies while trying to win your money is the best available forecast of how fast they’ll reply once they have it — and it will only get slower. Three cheap tests before you pay, each predicting a different failure.
The moment it happens
The late-night message test matters because responsiveness is the thing buyers most often complain about afterwards and least often check beforehand. Complaint records describe the same arc repeatedly: replies within minutes during the sale, then hours, then only after two follow-ups, then nothing.
That decay is predictable. The starting point is measurable before you commit.
Test 1: The odd-hour message
Send a real question at 9pm on a weekday, or on a Sunday afternoon. Not a trick — something you genuinely want answered:
Quick question — will the domain be registered in my name, and what’s the year-two cost for hosting?
Then note when the reply arrives.
| Reply timing | What it predicts |
|---|---|
| Within an hour, even at 9pm | Someone is actively engaged. Best available signal |
| Next morning | Perfectly reasonable. Normal working boundaries |
| After 24 hours | Slow now, slower later |
| Only after you follow up | This is your future, at its best |
| Never | Your answer, at no cost |
Don’t over-weight the late reply as a negative. A vendor with healthy working hours is not a bad vendor. What you’re looking for is the difference between “answers within a working day” and “answers when chased” — those two produce very different projects.
Test 2: The specific question
Send a question that requires actual thought rather than a template reply:
For a business like mine, would you recommend showing prices on the site or not?
This tests two things at once. Whether they answer the question, and whether they know your business well enough to have a view. A vendor who replies with a considered opinion — even one you disagree with — is engaging with your project. One who replies “yes sir, we will do as you like” has told you they’ll build whatever is asked and take no responsibility for whether it works.
The best possible response is a question back: “what do your competitors do, and are you cheaper or more expensive than them?” That’s someone scoping rather than selling.
Test 3: The small awkward request
Ask for something minor that costs them a little effort:
Could you send one client in a similar business I could speak to for two minutes?
Or: “Could you send me the year-two cost in writing?”
Watch what happens with a request that isn’t purely in their interest. Willingness here is the closest proxy available for how they’ll handle a small change request in month eight — which is exactly the interaction that decides whether you’re satisfied a year from now.
What each result actually predicts
Being precise about the limits of this:
Fast and specific replies predict a good build experience. They do not guarantee post-launch support, which has different economics — a vendor at the low end of the market genuinely cannot afford a year of small changes, however responsive they are now.
Slow replies before payment predict slower replies after, reliably. This is the strongest inference available from these tests.
Refusal on the reference or year-two cost predicts friction on anything that isn’t a sale.
So use the tests to eliminate, and settle support separately in writing rather than assuming responsiveness now will persist.
The thing to settle in writing regardless
Because responsiveness decays for structural reasons, not just personal ones, agree this before paying:
After launch, what’s your response time for a small change, and what does it cost? Is there a monthly plan?
Get it in a message. A vendor who names a response time and a rate has given you something enforceable. One who says “anytime, just message me” has given you goodwill, which is what evaporates in month eight.
Your own side of it
Worth saying, since this cuts both ways. Vendors deprioritise clients who are slow to reply, slow to approve, and slow to supply content. If you want to be the client whose messages get answered first, be the client who answers within a day and sends feedback as one numbered list.
That’s not a moral point. It’s the most effective lever you have over how you’re treated across a four-week project with five other clients in it.
What to do this week
- Send the odd-hour question to each shortlisted vendor and note the reply time.
- Send the specific question and read whether they engage or defer.
- Ask for a reference and the year-two cost, and watch the willingness.
- Get the post-launch response time and change rate in writing before paying.
- Commit to answering within a day yourself.
If you want it done the certain way
Test us before you pay — send an odd-hour question and see. We reply in about five minutes between 9am and 7pm, and we’ll put the post-launch response time and change rate in writing so it doesn’t rest on goodwill. WhatsApp us.
Related reading
- How fast should a website company reply?
- When replies go from minutes to days
- How often should a developer update you?
- Why support stops after full payment
FAQ
How can I test a web developer’s responsiveness before hiring?
Send a real question at an odd hour and note the reply time, send a question needing an actual opinion, and ask for something minor that costs them effort — such as a client reference or the year-two cost in writing.
How fast should a web company reply to an enquiry?
Within a working day at minimum, and often within an hour during business hours. What matters is the difference between “replies within a day” and “replies only when chased” — the second predicts a difficult project.
Does fast replies before payment mean good support afterwards?
Not reliably. Post-launch support has different economics, and at the low end of the market there’s no money in the price for a year of small changes. Get the response time and change rate in writing before you pay.
