Turn Customer Questions Into Page Improvements: Your First Month of Messages Is a Free Product Page Audit

Every question marks the spot where your page went silent

When questions pile up, it's tempting to do something big: a new layout, a longer description, warmer copy.

But warmer copy can't tell someone whether the jar is included. A sentence can.

Think about how the kit sells at the bakery counter. Someone picks it up, asks "is the jar in there?", and gets an answer in a second. Online, the buyer jumps between the photo, the description, the option picker and the delivery note. When a fact gets lost between those places, it turns into a message to you.

So a question isn't proof the page failed. It's a pin in a map, showing where a buyer had to stop.

Ask where the fact got lost

For each question, ask:

• Was the fact missing entirely?
• Was it there, but in the wrong place?
• Was it there, but in words the buyer doesn't use?

"Does the kit include the big jar?" turned out to be the second kind. The description said "dried starter in a glass jar". The photo showed a large 1-litre kilner jar used for feeding the starter, which isn't included. The fact existed. The photo contradicted it.

Start a question log, and copy every question word for word

You don't need anything fancy. A spreadsheet or a notebook works. For each question, write:

• The date
• The exact words of the question
• The product it's about
• Where it came from: email, message, comment, bakery counter
• Whether they'd ordered yet
• The answer you gave
• Where that answer came from: your guide, your recipe notes, your own testing

Copy the question exactly. Don't tidy it up.

"Is the starter alive when it arrives?" and "Does the starter need feeding straight away?" might look like the same question. They need different answers. The first is about what arrives. The second is about what to do with it.

And customers' words are often the words other buyers use when they search. "Starter smells like nail varnish" is something people genuinely type into search engines.

Note whether they'd ordered, because it changes which page needs fixing

This one column changes everything.

A question asked before ordering usually means the product page didn't answer it. A question asked after ordering might mean the product page, the printed guide or the follow-up email didn't answer it.

At Rising Hill:

• "Does the kit include the big jar?" came before ordering. That's a product page fix.
• "Is the nail varnish smell normal?" came after ordering, a few days into feeding. That's a guide and follow-up email fix, though it might deserve a line on the page too.
• "How long until I can bake?" came both before and after. Before ordering, it's about whether to buy now for a weekend bake. After, it's about planning.

Without the "ordered yet?" column, all three would have gone into one pile marked "product page problems".

Don't mistake one chatty customer for a trend

Once a few rows build up, patterns appear. Be careful with them.

Only mark a question as repeated when different people have asked it separately. Three questions from one enthusiastic customer aren't a trend. Neither is one person asking twice through email and a direct message.

The log isn't a survey. It's a way to spot missing facts without losing the detail of what people actually asked.

Sort every question into one of three piles before you touch the page

Not every question should change the product page.

Public facts. Answers that would help anyone: what's included, how long it takes, what it works with, what's normal. These usually belong on the page.

Guided choices. Answers that depend on the person: "Should I get the kit or just the starter?" depends on whether they already have flour and a jar. You can add a gentle tip, but with a clear condition.

One-off support. Personal situations: a parcel left at the wrong door, a cracked jar, a request to send a kit as a surprise gift. These belong in a private reply, not on the page.

The test for whether a question belongs on the page

Ask: is this about a fact that won't change, which another buyer could reasonably need before choosing?

"Does it include the big jar?" passes. "Can you post mine to my sister's address on Thursday?" doesn't.

This matters most in the first weeks, when a handful of messages can feel like a flood. Publishing an answer to every one-off request can make every future buyer expect the same.

Fix the riskiest gaps first, not the most frequent

How often a question appears isn't the only thing that matters.

For a food product, one question about allergens or safety outweighs five about the cloth cover's colour. Rising Hill had one message asking whether the flour was suitable for someone avoiding gluten. It isn't. That one message went to the top of the list, above the far more common jar question.

Ask of each gap: what happens if someone doesn't find this answer?

If the answer is "they might be disappointed", fix it soon. If it's "they might be harmed, or buy something unsuitable", fix it today.

Explain what's normal before it becomes a worried message

Some questions aren't about missing product facts. They're about what happens after, and the customer can't tell whether it's normal.

Sourdough starters go through stages that look and smell alarming to a beginner: a sharp, solvent-like smell when hungry, a layer of greyish liquid on top, bubbles that appear and disappear.

The "nail varnish" question came from three different customers in two weeks. So Rising Hill added a short "What's normal" section to the printed guide and the product page:

"In the first week, your starter may smell sharp, a bit like nail varnish, or develop a grey liquid on top. Both are normal and mean it's hungry. Stir the liquid in or pour it off, and feed it as usual. If you see pink, orange or fuzzy mould, throw it away and start again."

That last sentence is the important one. It tells people what isn't normal, so the reassurance doesn't hide a real problem. Only explain what's normal if your own testing and guidance back it up.

Put each answer right where the decision happens

Once you've chosen a public fact to add, put it where the buyer needs it:

• "What's included?": a clear list beside the main photo
• "How long until I can bake?": beside the button, because it affects when people order
• "Can I use a bread machine?": in the product details, near "what you'll need"
• "What's normal?": in the guide and lower on the page, for people already using it

A correct answer in an FAQ page technically answers the question. But if the buyer needs it before choosing, it's in the wrong place.

For the jar question, Rising Hill did two things: added "Feeding jar not included" as a caption on the photo, and added a "What's in the kit" list right beside it.

Follow the new fact through the basket and every email

When you add a fact, check it matches everywhere else:

• The option picker
• The basket line
• The confirmation email
• The printed guide in the box
• Your saved replies

The new page said "Feeding jar not included". The confirmation email still showed the old photo with the big jar. That would have created the same question all over again, from the opposite direction.

If two of your own records disagree, settle the real fact first. Don't polish the sentence until you know it's true.

Make one small edit, then check it

Before editing, write a three-line note:

• The question, in the customer's words
• The fact that answers it, and where that fact came from
• Where it will go on the page, and how you'll check it

Then make the smallest honest change that closes the gap. A one-line list of what's included beats a rewritten paragraph. A photo caption beats another adjective.

Check the live page afterwards. Read it as someone who's never seen it, find the new fact, and make sure it's where the question would arise.

Write down the old wording, the new wording, the source and the date.

If you can't confirm the answer, don't write around the gap

"Can I use the starter in a bread machine?" Rising Hill didn't know. They'd never tested it.

So the log said "need to test". Two of the bakers tried it the following week with the bakery's own bread machine. One recipe worked; one didn't. The page now says: "Our starter works in bread machines with a sourdough programme. Results vary between machines, so try a small loaf first."

A page that's honest about what it knows beats one that guesses.

One question can hide two decisions

"How long until I can bake?" sounds like one question. It's really two.

Before ordering, it means: "If I order today, can I bake for Saturday?" The answer depends on delivery time plus the starter's revival time.

After ordering, it means: "What's my schedule?" The answer is a day-by-day plan.

So the page got one line beside the button: "Allow about a week: 1 to 3 days for delivery, then 4 to 5 days to revive the starter before your first bake." The printed guide got the day-by-day plan.

When a question seems to need a long answer, check whether it's really two questions for two different moments.

A new question after a fix is progress, not failure

Fixing one gap often reveals the next. That's a good sign.

After the "what's included" list went up, the jar question stopped. A new one appeared: "What size jar do I need?" The buyer now understood the jar wasn't included, and moved on to the next decision.

That became one more line: "You'll need a clean jar of at least 750 ml."

Keep a "reply only" column too, for questions that depend on someone's personal situation. It stops the page from trying to answer every possible circumstance.

Write the answer the way you'd say it at the bakery counter

When adding an answer, there's a pull towards formal language: "Feeding vessel not supplied with kit contents."

Resist it. Think about how you'd answer face to face: "The big jar in the photo isn't included. You'll need your own, at least 750 ml."

The second version answers the actual question and sounds like a person who knows the product. It also uses the words buyers use, which makes the answer easier to spot when someone scans the page.

Feed the answers into the guide in the box, not just the page

Some questions belong on the product page. Others belong with the product itself, because they come up after it arrives.

Rising Hill's printed guide originally covered the feeding schedule and one recipe. After six weeks of questions, it gained three short sections:

• "What's normal in the first week": the smell, the grey liquid, the bubbles
• "What isn't normal": mould, and what to do
• "If your starter seems slow": warmth, feeding, patience

Each one came straight from the log. The next print run of the guide answered the questions before they were asked, and the follow-up emails could stay short.

Read the log before your next launch

The log is most valuable just before you launch something new.

When Rising Hill planned a rye starter kit, they read the whole log first. They already knew buyers would ask what's included, how long until baking, what's normal, and whether it works in a bread machine.

So the rye kit's page answered all four before it went live. The first month of rye questions was mostly about recipes, not basics.

What your question log can't tell you

Your log only shows people who got in touch. It doesn't show people who left without asking.

It can't tell you why someone didn't buy. A clearer page can't fix a price that's too high or bad timing.

And a quieter inbox doesn't prove you're selling more. It might mean a clearer page, or just a quiet week.

So keep conclusions small. "Three people asked about the jar, and after the caption went up nobody did" is accurate. "The caption increased sales" needs real evidence.

The questions nobody asks are worth a look too

Some gaps never produce a message. A buyer who's unsure about delivery to their area might just leave.

So once a month, read the page as a silent buyer. Look for facts that would make you hesitate but not quite ask: the delivery charge, what happens if the starter doesn't revive, whether it can be sent as a gift.

If you spot one, treat it like a question that arrived. Log it, check the source and add the answer beside the decision it affects.

Close every row with a status, so you never solve the same question twice

When you've dealt with a question, give it a status rather than deleting it:

• Published: the answer is now on the page
• Reply only: it belongs in personal replies
• Waiting for a source: you need to test or check first
• Left unknown: you've decided not to answer publicly for now

When the same question comes up after your next launch, you won't have to work it all out again.

If someone else helps answer messages, share the log with them, so everyone gives the same checked answer.

Your next step this week

Think of the last question you answered more than once. What's included, how long it takes, what's normal, or what it works with.

Check you have a reliable source for the answer.

Then add the answer to the page, right beside the choice it affects, in the words you'd use face to face. Check the basket and the confirmation email say the same.

One question turned into one clear fix. Keep the log open, and let the next real question show you what to fix next.

Keep going

Product Pages / Practical guide

Product Page Checklist: 12 Checks to Run Before Your Page Goes Live, Starting With the Fact That Rules Buyers Out

Most product page checklists count the parts of a page. These twelve checks test whether a buyer can understand, choose and order without guessing, using one made-up dog harness to show every check finding a real problem.

Product Pages / Practical guide

What Buyers Should See First on a Product Page: What It Is, Which Version, the Price and the Next Step, Before Any Story

A new visitor decides in a couple of seconds whether your page is worth scrolling. The top of the page doesn't need to say everything. It needs to say what the product is in ordinary words, which version is selected, what it costs, what to do next, and the one condition that could change the decision.