Facebook ads getting clicks but no sales: where to look

Check the campaign objective and measurement

Meta shows your ads to the people its system predicts are most likely to take the action you asked for. If you asked for clicks, you get people who click, and people who click are not always people who buy.

• The objective.

  Meta offers campaign objectives aimed at outcomes such as traffic, engagement and sales. A traffic objective looks for people likely to click, which can mean cheap clicks from people with little intent to buy. For an online store, a sales objective optimised for purchases is usually the relevant comparison, provided purchase tracking works.

• The optimisation event.

  Within a sales campaign, check which event it is optimising for. A campaign optimised for add to cart or landing page views will find more of those, not necessarily more purchases.

• The events your site sends back.

  Meta learns from signals your store sends through the Meta Pixel and, where it is set up, the Conversions API. If purchase events are missing, counted twice or firing on the wrong page, the campaign is optimising blind.

• Which clicks you are reading.

  Meta reports more than one click measure. Broad click counts can include taps on your page name, on ‘see more’ or on the image. Link clicks are closer to what you want, and landing page views are closer still, because they count people who stayed until your page loaded.

Use Meta’s Events Manager, or whatever currently replaces it, to check that purchase events arrive with the right values. Then compare the purchases Meta reports with the orders in your store for the same dates. The two will not match exactly, because they count differently. But a campaign showing no purchases while your store shows orders from Meta traffic, or the reverse, is a tracking problem to fix before anything else.

If a campaign is optimised for clicks and judged on sales, it is being measured against a goal it was never given.

Follow the click through the purchase journey

Now take the path a buyer takes. Open the ad on your own phone inside the Facebook or Instagram app, tap through, and try to buy.

1. 1

   The tap

   Links from Facebook and Instagram often open in the app’s own browser. Saved passwords, autofill and some payment options can behave differently there. Check that your pages load and your checkout works inside that browser, not only in your usual one.

2. 2

   The landing

   Compare link clicks with landing page views. A large gap means people tap and leave before the page appears. Page speed on mobile data is the usual suspect, followed by broken or redirected links.

3. 3

   The first screen

   Does the page show the product from the ad, at the price the ad implied, on the first screen of a phone? Sending ad traffic to a homepage or a broad collection makes people hunt for what they have just seen, and many will not.

4. 4

   The product page

   Can a shopper answer their questions about size, material, how it works, delivery time and returns without leaving the page? How to improve product pages for real shoppers covers what the page needs.

5. 5

   The cart and checkout

   Any surprise here, such as shipping costs appearing for the first time, a forced account or an error on mobile, undoes the work the ad did.

Then read the journey in numbers for this campaign only. Your analytics tool, filtered by the campaign’s tagged links, and Meta’s own reporting can show how many people reached each step. The step with the steepest drop is where to look first. Keep the numbers in proportion: with a few dozen visitors, one or two people can make a step look far better or worse than it is.

Test the match between creative and offer

An ad makes a promise and the store has to keep it. Many campaigns with clicks and no sales have ads that are very good at attracting attention and not specific enough about what is being sold.

• Curiosity clicks.

  An intriguing video that reveals the product only at the end collects clicks from people who wanted to see what happens, not what it costs.

• Discount-led clicks.

  A big headline discount attracts bargain hunters. If the page then shows the discount only applies to some products or above a minimum spend, they leave.

• Interest without need.

  The ad reaches people who like the look but are not in the market, such as gift-style products shown to people who are not buying gifts right now.

• An offer that cannot pay for the click.

  Sometimes the ad and the page are fine and the offer is the problem. If the average order is small and the margin thin, even a healthy share of buyers cannot cover what a customer costs to win.

The last one is worth doing the arithmetic on, because it looks exactly like an ad problem from the inside.

• $3,204 ad spend in January 2026, Apothecary by Justin
• $548 revenue from 8 orders that month
• about $400 to buy a customer
• about $55 gross profit per customer at a $69 average order

The ads were not what was broken. There was no realistic cost at which a customer worth about $55 could be bought. Once the average order rose from $69 to $137, the same kind of advertising could pay. The case study has the full sequence, and what a good ROAS is for your business shows how to work out the return your own costs require.

To test creative and offer separately, change one side at a time. Keep the offer fixed and test two or three angles that describe the product plainly, showing it early. Or keep the creative fixed and test the offer, such as a bundle that raises the average order. Changing both at once tells you nothing about which one mattered.

Decide what evidence to collect next

The question founders ask most at this point is how many clicks, or how many days, before they know. There is no honest universal number. It depends on your price, your margin, what a sale is worth to you and how much you can afford to spend finding out.

What you can do is set the rule before you spend, in your own numbers.

With that in hand, choose one next move: fix tracking, fix a specific step in the journey, test a different angle, test the offer, or stop and put the money somewhere else. Stopping is a legitimate result. So is discovering that the platform’s figures and your store’s disagree, and deciding which to trust. Revenue isn’t proof it’s working covers that problem.

If you would rather hand this to someone, our Meta ads work starts from the business behind the campaign rather than the ad account. The resource hub has more guides on ads, measurement and conversion.