Small Shop Notes

Why Your Shop Isn't Growing (Even When You Have Products)

Why Your Shop Isn't Growing (Even When You Have Products)

About Alex: a pen name. There is no shop behind it, and no results are claimed. What follows is arithmetic and opinion, and I'll mark which is which.

In short

Before you make anything new, find out which of four stages in your shop leaks: Reach, Interest, Purchase or Profit. Then work out what one sale leaves you, because that profit per order decides what any extra traffic is worth. (That order of work is my opinion; it is also the cheaper one.)

What you get from reading it:

  • A 7-number snapshot of your last 30 days
  • Your profit per order
  • One named leak and one written test to start this week

You made the thing. It took weeks, maybe, and now it sits in your shop looking finished, and nothing happens. So you open the stats page. A few visits, a few orders, a monthly total that doesn't match the evenings you put in. You close the tab, and ten minutes later you open it again, as if the number might have changed its mind.

Then comes the tempting thought: make something new. A bundle, a fresh listing, product number twelve. Or new photos for everything, or a sale, or a different business altogether.

I'd hold off, and the reason is cost. Making something new is the expensive way to learn what's wrong, because you spend a week of evenings and only then find out whether the product was ever the problem. The question I'd ask first is plainer: what exactly isn't working with the products you already have?

"Not selling" hides at least three different situations, and they need different responses.

Three situations, three responses. Before you pay for another tool or subscription, I'd look at the numbers you already have. Paying for a fix before you understand the problem is the other expensive way to learn what's wrong.

I'll call these three places, plus one more, leaks. A leak is a stage on the path from a stranger to your profit where more people, or more money, disappear than you'd like. "My first image doesn't answer a question" is something you can fix on a Tuesday evening. "My shop isn't growing" gives you nowhere to start.

Who this is for. Sellers with at least one product on sale, on a marketplace like Etsy or on their own site. If you have nothing listed yet, a different starting point will serve you better. If your shop is already comfortably profitable and growing, skim Step 6 and carry on with your life.

What it costs. Nothing, and no new software. I'm wary of anything that opens with "first, sign up for a tool"; a spreadsheet or a notebook is enough. You need your shop's stats and one sitting for the snapshot. The Step 5 test then runs a week or longer.

You'll finish with a one-page Shop Snapshot, a profit-per-order number, one named weak spot and one test to start this week.

The traffic question in one line: orders you need = monthly profit target ÷ profit per order, and visits you need = orders you need ÷ your visit-to-order rate. Step 2 puts numbers on it.

One habit keeps the whole exercise honest. Write the number you read from the dashboard, and use 0 only when the dashboard really shows zero. If nothing in your setup counts that thing, write not measured. If a counter exists but shows nothing, write unknown.

Mixing these up turns "I never installed anything that counts sign-ups" into "I get zero email sign-ups". One sends you off to fix a problem you may not have; the other tells you what to start counting. (The weekly routine in A Fifteen-Minute Weekly Review for a Small Shop shows how to keep a log of the numbers you do count.)

Three imaginary sellers carry the examples. Maya sells budget spreadsheet templates, Tom printable wall art, Priya digital stickers. None of them is a real person or a real shop, and every number attached to them is invented to show the arithmetic.

Step 1: Put your 30-day numbers on one page

You'll end up with seven numbers, each carrying the right label.

"My shop is stuck" gives you nothing to work with. "4,800 visits and 24 orders" does. Everything after this is arithmetic on these seven numbers, so get them right once.

Open your shop's stats. On Etsy.com, Etsy's help page describes the route as Shop Manager > Stats (as of 25 October 2026; menus do move, so if yours looks different, search the help centre for "Stats"). The page shows visits, orders, conversion rate and revenue, with a date drop-down.

Pick the last 30 days, so a quiet weekend and a busy one even out. Other marketplaces and website platforms have something similar under other names. Look for a date range and the terms below.

Etsy's help page says visit data refreshes a few times a day to filter out bot traffic, and the count may be lowered again at the end of the day; listing views update in real time (as the page reads on 25 October 2026). So a visit count that hasn't moved in ten minutes isn't broken.

Four words you'll meet. Etsy's glossary describes the key terms like this [S37]:

If you sell elsewhere, don't assume the same definitions. Find the equivalent terms and write down what your dashboard says each one means. The differences matter later.

For revenue, keep it simple. If the dashboard's definition confuses you, add up the sale prices of your last ten orders and divide by ten. That average is all Step 2 needs. Etsy's revenue figure leaves out shipping and tax, so count shipping the same way in rows 4 and 5, either out of both or in both.

Here is the snapshot to fill in.

SHOP SNAPSHOT (last 30 days)
From ____ to ____
1. Visits: ____  state: ____
2. Listing views: ____  state: ____
3. Orders: ____  state: ____
4. Avg order value: $ ____
   (revenue ÷ orders, or last-10 avg)
5. Fees and costs per order: $ ____
   (fees, ads, materials, delivery)
6. Email sign-ups: ____ state: ____
7. Pinterest visits: ____ state: ____

Fill-in snapshot sheet

Rows you cannot fill stay as "not measured" or "unknown". Use 0 only when the dashboard really shows 0.

1. Visits
2. Listing views
3. Orders
4. Average order value ($)—
5. Fees and costs per order ($)—
6. Email sign-ups from your shop
7. Visits from Pinterest
Automatic arithmetic (stays on this screen; nothing is sent or saved)
  • Profit per order: —
  • Visit-to-order rate (orders ÷ visits): —
  • Orders needed: —
  • Visits needed (at your current rate): —
  • Your visits are —
  • Views per visit: —

Figure: the snapshot as a one-page sheet. The plain-text version to copy is the box above; Maya's filled-in version comes next.

"State" is where you note which kind of number it is: read (you saw it on the dashboard), not measured, or unknown.

Row 5 is the one I'd be tempted to skip, because it's the row where the number gets smaller. Look at the fee and payment lines on a few recent orders (your marketplace may list them on its order or payment pages; I haven't checked) and average what was taken per order.

Ads, materials and delivery go in row 5 too. Write the real number even if it stings. A profit figure that leaves costs out flatters you, and then you plan around money that never arrives.

If you sell only on a marketplace, row 6 will probably read not measured, because a marketplace doesn't hand you a sign-up form. That's fine. Write it down and move on; Step 6 picks it up.

Row 7 has a trap, in two parts. On Etsy, look in the stats section called "How shoppers found you". Etsy's glossary describes a Social Media traffic source covering Instagram, Facebook, Pinterest and other sites, so Pinterest may be lumped in with the others. I haven't seen your dashboard, so read what it actually shows, and write unknown if Pinterest isn't shown on its own.

On the Pinterest side, an outbound click is counted when someone takes an action that sends them from a Pin to a destination off Pinterest. A Pin click is a different thing: it counts the Pin being opened up close [S34]. A Pin can have plenty of Pin clicks and few outbound clicks.

Outbound clicks aren't the same as your shop's visits either: some people may click and leave before the page loads, and some visits may not be recorded. Pinterest's help page also says outbound clicks are measured from the date the link was added, which can explain a mismatch for new links. Record both numbers, note the gap, and resist the urge to make them match.

Here is Maya's page after one sitting, so you can see the target.

SHOP SNAPSHOT (last 30 days)
1 Sep to 30 Sep -- Maya, imaginary
1. Visits: 1,240  state: read
2. Listing views: 3,100  state: read
3. Orders: 31  state: read
4. Avg order value: $18.00
   (avg of her last ten orders)
5. Fees and costs per order: $4.50
6. Email sign-ups: --
   state: not measured
7. Pinterest visits: --
   state: unknown

Two rows hold a label instead of a number, which is perfectly good: Maya now knows she has no sign-up counter and hasn't looked at where her visits come from. The page takes ten seconds to read, and you'll come back to it in Step 3. If a number won't turn up, write not measured and keep going.

Sources: [S34] [S37]. Hands-on verification: pending (V01, V02).

Step 2: Find out what one sale really leaves you

You'll end up with a profit-per-order figure, and a calculation that runs backwards from a target.

Revenue is what customers pay. Profit per order is what you keep after the cost of making that one sale. A shop can have plenty of orders and still be a poor place to spend your evenings if each one leaves a few cents.

I'd work this out before chasing any traffic, because traffic multiplies whatever number sits underneath it.

Take your average order value (row 4) and subtract the costs of that sale (row 5, split out here):

Average order value:    $ ____
 − fees (shop, payment): $ ____
 − ads per order:        $ ____
 − materials/delivery:   $ ____
 = Profit per order:     $ ____

Now the backwards part. Start from a goal: choose a monthly profit target. Divide it by your profit per order to get the orders you need. Divide that by your visit-to-order rate (orders ÷ visits) to get the visits you need.

None of this is a forecast. It only measures how far your present shop is from the thing you say you want.

Maya, from Step 1, sells a spreadsheet template for an average of $18 an order. After $4.50 of fees and costs, her profit per order is $13.50.

Say she imagines a target of $1,000 profit a month. That's a practice target for the arithmetic. It isn't a prediction, a typical result or a promise, and this site has not achieved it.

So her listings already turn about one visitor in 40 into an order. A new listing might add some visits, but she'd need a lot of them to close a gap of about 1,760 visits a month.

What the numbers don't say: that 2.5% is good or bad, or that 3,000 visits would arrive or convert the same way. They show how far her current visits are from her own target, and that's all.

Three snags:

Sources: none; this step is arithmetic. Hands-on verification: not applicable.

Step 3: Name the stage where your shop leaks

You'll end up naming the stage where your shop is weakest, using your own numbers.

Think of the path from a stranger to your profit as four stages, with a possible leak at each. The table gives the one number I'd look at for each.

Stage Question Look at A leak looks like
1. Reach Do enough people arrive? Visits vs the Step 2 target Visits far below target
2. Interest Do visitors look further? Views ÷ visits Many visits, almost no views
3. Purchase Do interested visitors buy? Orders ÷ visits Many visits, very few orders
4. Profit Does a sale leave enough? Profit per order Many orders, thin profit

Maya's 3,100 views over 1,240 visits is 2.5 listings per visit, so her visitors do look around. Interest isn't her trouble.

I have no trustworthy outside "normal" to hold your numbers against, so the table works with a rule and your own figures.

A lever is one thing you could change to move your numbers: more visits, a better page, a different price.

Why pull one lever first, and how to pick it

A rule you don't understand is easy to abandon, so the reasoning first. You have a limited number of evenings and more than one thing that could be better. Two things should decide what goes first: whether the change could matter, given what a sale leaves you, and how cheaply you can try it.

Reach and purchase look alike in the arithmetic. Say Maya needs 75 orders and has 31. That's 75 ÷ 31 = 2.4 times as many. She could get there with 2.4 times the visits, or a visit-to-order rate 2.4 times higher, or some mixture. The arithmetic treats both the same and can't choose for you.

What can choose is the cost of trying. A page change takes one evening, and you read its effect in numbers you already collect (Step 5 says how many visits that needs). I'd guess adding a lot of traffic is slower and less in your control.

If you have a cheap, reliable source of visitors, that may not hold for you, so treat the rule below as a default.

A rule for picking the first lever (a judgement, not a benchmark):

  1. Profit first. If a sale leaves less than you'd accept, that's the leak. More visits would only multiply a thin margin. Stop here.
  2. Otherwise let Step 4 choose. Do the "clicks but no sales" check on your own listing. If it turns up open questions, test the page first. If the page already answers what visitors need, Reach is your lever.
  3. Interest is the odd one out. If listing views per visit is very low (imagine 1,000 visits and 300 listing views, so most visits never open a listing), the front of the shop or the traffic source is the likely issue, and Step 4 applies there too. None of the three shops below is an Interest case.

Here are the three imaginary shops side by side. The $1,000 row is the same practice target as before, not a promise.

Maya (spreadsheets) Tom (wall art) Priya (stickers)
Visits 1,240 4,800 2,000
Orders 31 24 50
Visit-to-order rate 2.5% 0.5% 2.5%
Average order $18.00 $18.00 $9.00
Costs per order $4.50 $4.50 $6.50
Profit per order $13.50 $13.50 $2.50
Profit in 30 days $418.50 $324.00 $125.00
Orders needed for a $1,000 target 75 75 400
Visits needed at own rate 3,000 15,000 16,000
First lever (judgement) 1. Reach 3. Purchase 4. Profit
Maya
41.3%
Visits today: 1,240 / Visits needed: 3,000
Profit per order: $13.50 · Visit-to-order rate: 2.5%
First lever to test: Reach
Tom
32.0%
Visits today: 4,800 / Visits needed: 15,000
Profit per order: $13.50 · Visit-to-order rate: 0.5%
First lever to test: Purchase
Priya
12.5%
Visits today: 2,000 / Visits needed: 16,000
Profit per order: $2.50 · Visit-to-order rate: 2.5%
First lever to test: Profit
Figure 2. Three imaginary shops: today's visits as a share of the visits a $1,000/month profit target would need (all invented practice numbers)

"I need more sales" can mean three different jobs, and the table sorts them.

Tom works it out slowly, wrong turn included

The table shows only the answer. Here is Tom getting there, wrong turn included.

His snapshot says 4,800 visits and 24 orders. He reads the second number twice, and his first guess arrives: the price is too high. People come, look, and leave, so surely they find it expensive. He thinks about dropping his $18 print to $15.

Before he does, he runs Step 2 on the idea, treating his costs as unchanged:

The cut doesn't close his gap. It makes the climb steeper, unless the lower price brings in a wave of extra buyers. His first guess had a cost he hadn't seen. He leaves the price alone and notes it as a possibility for later.

The other option is more traffic. His numbers: 15,000 visits needed against 4,800 today, a gap of about 10,200. That gap is about 2.1 times the traffic he has now, so he'd need about 3.1 times today's visits in total, and he doesn't know how he'd get them.

The 0.5% beside it is worth a look too: out of every 1,000 visitors, about 5 buy.

Turn it around. If traffic stayed exactly where it is, what rate would 75 orders need? 75 ÷ 4,800 = about 1.56%, roughly three times his 0.5%. Also a large change.

Neither lever is small, so which is cheaper to test? Changing the first image of one listing takes an evening and costs nothing, and the answer arrives in numbers he already collects. Adding 10,000 visits is a bigger, less certain job.

Nothing in the numbers proves the page is the problem; they show only that testing the page is the cheapest way to find out. On to Step 4.

For your own shop, name the one weakest stage and the two or three numbers that point to it. If two levers look equally believable, pick the one that costs least to test this week.

Sources: none; this step is a method with invented example numbers. Hands-on verification: not applicable.

Step 4: Work out why clicks are not turning into sales

If your leak is at Purchase or Interest, this step helps you find which of three causes is the likeliest. I'd do it even if you think the leak is Reach, because it's how you rule the page out.

This is Tom's situation: people are arriving, so why aren't they buying? A listing that gets opened and abandoned leaves you guessing about everything, and nobody tells you which part. Before you change anything, check three things in order, all free.

1. Does the page answer what the visitor came for? If your traffic comes from a Pin, read the Pin as a stranger would, then read the first screen of the page it links to.

A Pin promising "three ways to organise a monthly budget" that lands on a product page with no mention of budgeting has broken its promise. The visitor leaves, and your dashboard records a visit with no order. (If you promote through Pinterest, this guide, Make Your Pinterest Click Land on the Right Page shows how to check the match.)

2. Is it the wrong crowd? Some traffic simply isn't shopping. Someone who clicked a free tip isn't necessarily looking to buy a template today. If most of your visits come from one source, split your numbers by source for a week and compare how each group behaves.

A Pin-specific tracking tag helps: a small label added to a link so your stats can tell the sources apart. The guide to matching a Pin to its page covers tagging.

3. Is a small question blocking the sale? Imagine a visitor who likes your planner thumbnail and pauses. Can I fill it in on my phone? What would my first Monday look like? Is it editable? Those questions never reach you. The page either answers them or it doesn't.

Read your own listing as a first-time buyer and list every question it leaves open.

Tom does exactly that with his wall-art listing and finds three: "What size can I print this at?", "Will it look right in a frame I already own?" and "Does it come in A4 as well as US Letter?" The first screen shows one photo and a title, with sizes buried under "more".

From that he writes a hypothesis, which just means a guess stated clearly enough to be proven wrong: showing the print sizes on a framed example in the first image will raise the visit-to-order rate. It's testable. It isn't a fact.

Before you leave this step, write down at least three concrete questions a first-time visitor might still have, labelled as guesses, and one source-split you could run. If every question seems answered, the issue could be price, trust (few reviews) or the traffic itself; test the cheapest of those in Step 5.

Sources: none. Hands-on verification: not applicable (reading your own listing).

Step 5: Write one test you can run this week

You'll write a test small enough to start this week and honest enough to learn from. The window is seven days if your traffic allows, longer if it doesn't.

Change one thing only. If you change the photo, the title, the price and the description at once and sales rise, you won't know which change did it, and if they fall you won't know which to undo. By the fourth evening of watching an unchanged number, changing everything starts to look attractive. Resist it.

Write the test down before you begin, so the result can't be bent afterwards to look like what you hoped for:

LEAK: (stage and the numbers that
  point to it)
HYPOTHESIS: If I change ____, then
  ____ will improve, because ____.
WHAT I CHANGE: (exactly one thing)
HOW I MEASURE: (one number, as "out
  of how many visits", same dashboard)
START / END: ____ to ____
  (fixed before you start)
RESULT THAT WOULD CHANGE MY MIND:
  ____

Tom's version:

LEAK: Purchase. 4,800 visits,
  24 orders, 0.5%.
HYPOTHESIS: If I show the print
  sizes in the first image, more
  visits will end in an order,
  because buyers can answer "will
  it fit my wall?" without scrolling.
WHAT I CHANGE: The first image
  only (sizes on a framed print).
HOW I MEASURE: Orders out of
  visits, for the test window.
START / END: 3 Nov to 30 Nov
  (four weeks, fixed in advance).
RESULT THAT WOULD CHANGE MY MIND:
  Fewer than 30% more orders per
  visit than the four weeks before.

Why four weeks and not seven days? Tom's baseline, the "before" number he compares against, is about 5 orders per 1,000 visits. In one week he'd see roughly five or six orders, and 5 against 8 is well inside the range of chance.

So he commits in advance to four weeks (about 4,500 visits, roughly 22 orders at his current rate) and to a bar he sets before looking, meaning a threshold: at least 30% more orders per visit. The bar is arbitrary. What matters is that it's written down first.

At Tom's rate, 30% more orders is about 29 orders instead of 22, a gap of 7. Chance alone can move a count of 22 by several orders, so even a pass is only a reason to run another four weeks, not proof.

Small numbers deserve honesty. A rate from a small sample (the handful of visits you happened to get) is mostly noise. If 1 visit in 50 ends in an order (2%), the true rate could plausibly be anywhere from about 0.35% to 10.5%. That range is wide enough to hide almost any improvement you could make.

So if your shop gets under about 200 visits a month, a conversion test can't tell you much. I'd treat that week as a reading-and-fixing exercise: make the change, watch an earlier number such as listing views per visit, and come back when you have more visits.

Mark any result inconclusive rather than declaring victory or defeat. It's not a satisfying word, but it's an accurate one. The weekly review guide has a small-counts section and a keep/change/wait table.

The test is ready when it names one change, one measure written as "out of how many visits", a date range and a result that would change your mind. If you keep wanting to change two things, write the second down as the next test and don't run both.

Sources: none; this step is a method. Hands-on verification: pending (V03).

Step 6: Decide whether an email list is your next lever

An email list here means people who have said, in their own words, that they'd like to hear from you again. It's a permission-based way to keep helping someone who already showed interest.

"Build a list" is not the answer to everything. If your problem is a thin profit per order (Priya), a list doesn't change the arithmetic. If it's the page (Tom), fix the page first. A list won't create demand by itself, and not every subscriber becomes a buyer.

It earns its place when your leak is Reach: when you keep paying in time or money to find strangers who might have been glad to hear from you again. Maya is that case.

Pinterest and search can bring first visits, but a visitor who isn't ready today is gone unless there's a reason to stay in touch. One helpful, free next step (a useful article, a sample, a short course) gives her a way to keep that visitor.

If you sell only on a marketplace, you have no page of your own for a sign-up form yet; the next-step section below points to the guides for that. For now, just note whether your answer is "yes, a list could help".

One caution for marketplace sellers. I have only read Etsy's rules on marketing email as a search excerpt, not the full text, so check the current rules yourself.

As the excerpt reads, you need a buyer's express consent before sending marketing email, a sale doesn't count as consent, addresses received through a transaction may only be used for Etsy-related communications, and you may not add an Etsy member to a mailing list without consent [S33]. A sign-up form on your own page, separate from the purchase, is the route those rules point to. How to Grow Your First Email List as a Product Seller is about doing that properly.

One sentence will do: is your current leak one an email list could plausibly help, and why? If you aren't sure, run the Step 5 test first and come back in a week. I don't know which way your answer will go, and nobody reading only your numbers could either.

Sources: [S33]. Hands-on verification: not applicable.

Troubleshooting: when the numbers look wrong

Symptom Likely cause What to do
Your dashboard has no 30-day option It shows other ranges Use the closest range and write the dates at the top
Visits and Pin outbound clicks don't match Possible causes: click-and-leave, ad blockers, unrecorded visits Record both, write the gap, don't force it to zero
Today's visit count hasn't moved Etsy says visit data refreshes a few times a day, not live Check once or twice a day; the timestamp on the stats page shows when it last updated
Conversion rate on a non-Etsy dashboard differs from orders ÷ visits Other platforms may define it their own way Read that platform's help page, or calculate your own from orders and visits and say so
Profit per order is negative Costs or ads exceed price Fix price or costs before seeking traffic
Every stage looks fine and sales are still flat Not enough data, or a seasonal dip Extend the window; compare the same period last year if you have it
You change three things "just to be safe" Understandable impatience Revert to one change; log the others for later tests

Frequently asked questions

How many visits do I need before the numbers mean anything? There isn't one number. As an illustration, with a larger rate than any shop in this guide has: out of 100 visits, a measured 10% could be anywhere from roughly 5.5% to 17%; out of 1,000 visits it narrows to about 8% to 12%. For small rates you need many more visits for the same precision. The longer guide on reading numbers gives the table.

What's a good conversion rate? I can't tell you, and this guide gives no benchmark because none was verified from a source. Judge your rate against your own earlier rate, and against what your Step 2 maths says you need.

Should I lower my price? Not before you know your profit per order and your leak. A lower price helps a Purchase leak only if price is the reason people leave, and it hurts a thin-profit shop. Tom's walk-through in Step 3 shows the arithmetic.

I don't sell on Etsy. Does this still work? Yes. Replace the dashboard terms with your own platform's equivalents and note what each one means in its help pages.

Do I need to start an email list? Only if your leak is one it can help (Step 6). Otherwise fix the page or the price first.

What if my shop is brand new and has almost no visits? Then the first problem is Reach. Make the few visits you do get count by checking the page against the Pin or link that sent them (Step 4), and read the next guide on building an audience.

And if the stats page has been quiet all week, I won't tell you it doesn't matter. It just isn't a verdict on the product, because with this few visits you can't tell yet.

Where you are now: what you should be holding

If you did the steps, you have a profit-per-order number, one named weak spot and a written test. That's enough to start on Monday. You're done when every snapshot row holds a number or a label, profit per order is calculated, and your test names one change, one measure, dates and a result that would change your mind. And if you only read this far and come back later with your own numbers, nothing is lost.

Somewhere in the test window an order may arrive from a stranger who decided the page was for them. I'd let that be a small good moment before you check what it does to the percentage.

When you run the test, record the result as "out of how many visits".

Your next step: choose by where your shop leaks

Match the line to the leak you named in Step 3. None of these is required, and the honest answer for some shops is "no new tool".

If your leak is not Reach, you do not need an email tool yet; stop here.

If you sell on Etsy and want the consent rules laid out before you decide, Build a Consent-Based Email Course Outside Etsy is the one to read.

Verification status

How we checked. On 25 October 2026 we read Etsy's Shop Stats glossary and its "How to Use Etsy Stats" help page in full, and Pinterest's Business help page on reviewing analytics. That covers the term definitions, the Shop Manager > Stats location, the refresh behaviour, and the Pin click and outbound click wording. The Etsy rule on marketing email consent [S33] was read only as a search excerpt on 24 October 2026; according to the search excerpt we read, full text unchecked. Nothing here has been tested in a live shop.

Sources, as checked.

Checks still to run. Until real sellers have filled in the snapshot and run a test, the article stays in review.

What this guide does not claim. That any conversion rate is normal, that any of the imaginary shops existed, or that following it will raise sales.