If you are asking “why are my Facebook ads not converting?”, the money is already gone and the answer is somewhere in a chain of six links. The ad has to be delivered, the conversion has to be counted, the goal has to be set correctly, the audience has to be capable of buying, the click has to land somewhere that works, and what happens after the lead has to close.
A break anywhere in that chain looks identical in Ads Manager: spend going out, conversions not coming back. This article takes the six in the order that costs you the least time to check, so you stop at the first real fault instead of rebuilding a campaign that was never the problem.
Where Can Conversion Break?
Conversion can break in six places, and only the first three are inside Ads Manager. Delivery can stall so the ad barely runs. Measurement can miss conversions that did happen. The optimisation goal can be pointed at the wrong action. The audience can be people who were never going to buy. The click can land on a page that fails. And the lead can die after it reaches you.
Those six are ordered on purpose. Each one, if broken, makes everything below it unreadable. There is no point judging your audience quality while half your conversions are going uncounted, and no point rewriting creative while the ad set is spending 20 EGP a day and never leaving the learning phase.
Conversion: the specific action you asked Meta to optimise for — a purchase, a lead form submission, a message, a call — not a like or a click.
Check If Your Ads Are Actually Broken Before Fixing It
Before changing anything, establish that there is a fault to fix. Three checks take about ten minutes and settle it.
Check the window, not the day. Look at a period long enough to contain results, not yesterday. An ad set that gets a handful of conversions a week will show zeros on plenty of individual days without anything being wrong.
Check whether the ad set ever stabilised. Meta states that ad sets exit the learning phase after about 50 results in the week after the last significant edit, and that during the learning phase ad sets are less stable and usually have a higher cost per action. An ad set that never reached that point has not underperformed. It has not finished starting.
Check what changed, and when. Meta’s own delivery troubleshooting tells you to use breakdowns to see how results changed over time, then check the campaign’s activity history for edits made just before delivery slowed. Frequent pauses, budget changes late in the day and schedule extensions are all named by Meta as things that disrupt pacing.
If all three come back clean and conversions are still absent, you have a real fault. Work down the list.
What to Check, And in What Order
1: Delivery – Is the Ad Even Running?
Start here because it is the cheapest thing to rule out and the most common thing to be wrong. An ad that is not delivering cannot convert, and Ads Manager reports a non-delivering ad and a badly performing ad in the same column.
Three states are worth separating. The ad is still in review. Meta reviews ads against its Advertising Standards before they run, and states that most ads are reviewed within 24 hours, though it can take longer and ads can be re-reviewed after going live. The ad was rejected, in which case the Delivery column says so and the fix is the ad, not the budget. Or the ad is technically active and barely spending, which is the case the next two sections cover.
One number worth knowing before you panic about daily spend: Meta states it may spend up to 75% over or below your daily budget on a given day. A day that looks like under-delivery can be inside Meta’s stated pacing range.
Still in the Learning Phase
The learning phase is the period when Meta’s delivery system is still working out how to deliver an ad set, and it is the single most common reason an ad spends without converting. Meta states an ad set exits it after about 50 results in the week following the last significant edit, and that ad sets are less stable and usually carry a higher cost per action while inside it.
Two consequences follow. If your budget cannot buy roughly 50 of your chosen conversion in a week, the ad set will not exit, and its costs will stay unstable for as long as it runs. And every meaningful edit resets the clock, so an account edited every few days never leaves the phase at all. Meta’s own guidance is to wait until an ad set is out of the learning phase before editing it, and to avoid unnecessary edits that push it back in.
The arithmetic is simple enough to run now. Fifty results a week, at your current cost per result, is the weekly budget the ad set needs before its numbers mean anything.
The Auction Got Expensive (Ramadan and Seasonal Spikes)
Nothing about your ad has to change for it to stop working. When more advertisers bid for the same people in the same weeks, the same budget buys fewer impressions, and fewer impressions produce fewer conversions. Ramadan is the clearest example in Egypt, and the seasonal peaks in your own category do the same thing on a smaller scale.
Meta names the mechanism directly in its delivery troubleshooting: if other advertisers are bidding more aggressively, your cost per result goal or bid cap may not be high enough to stay competitive, and the options are to raise or remove the cap, or move to an optimisation event higher up the funnel.
This is the diagnosis behind “it worked last month and stopped this month” when nothing in the account changed. What the auction costs, and what sets that price, belongs to the Facebook advertising cost pillar rather than this article.
2: Measurement – Conversions Happening, Just Not Counted
Some campaigns are converting and reporting zero. The sale happened; the signal never arrived. This is worth checking second because every judgement below it depends on the numbers being real.
Two mechanics cause most of it.
The attribution setting decides what gets credited. Meta’s standard attribution settings count click-through conversions within 1 day or 7 days of a link click, view-through conversions within 1 day of an impression, and engage-through conversions within 1 day of a non-link click. A purchase that happens outside the window you selected is a real sale that your ad will never be credited for. Longer consideration cycles lose the most this way.
Browser-side tracking drops events. Meta states that data sent through the Conversions API is less affected than the Pixel by browser loading errors, connectivity problems and ad blockers, and that running both together makes the connection more reliable and helps the delivery system lower cost per action. If you are running the Pixel alone, some share of your conversions is not reaching Meta — and Meta cannot optimise toward events it never received.
The second point matters twice over. Missing events do not only distort your report. They starve the optimisation you are about to blame in the next section.
3: Optimization – Are You Telling Meta the Wrong Goal
Meta optimises for exactly the event you selected, and it is very good at it. An ad set optimised for link clicks will find people who click and stop there. An ad set optimised for landing page views will buy views. Neither is broken when it fails to produce sales; both are doing the job they were given.
Meta states that some optimisation events require more budget than others depending on how hard they are to achieve, and that optimising for purchases may need more budget than optimising for landing page views. That is the trade. The event closest to revenue gives the delivery system the best signal and costs the most to buy enough of.
So there are two failure modes, opposite to each other. Optimising for a cheap event and expecting sales, or optimising for a rare event on a budget that cannot buy 50 of them a week. The first produces traffic that never buys. The second produces an ad set stuck in learning.
4: Audience – Reaching People Who Were Never Going to Buy
If delivery, measurement and the goal all check out, the next question is who the ad reached. Meta’s delivery troubleshooting names two audience-side faults explicitly.
Audience saturation: a high percentage of your audience has seen your ads without acting. Meta’s stated remedy is to increase audience size.
Creative fatigue: the audience has seen the same creative too many times, and Meta shows Creative limited or Creative fatigue in the Delivery column when it detects this.
There is a size problem in both directions. Too narrow and the ad set cannot find 50 conversions a week, keeping it in the learning phase — Meta lists small audience size among the reasons an ad set becomes Learning limited. Too broad and the budget spreads across people with no intent. Facebook ads audience size is not a setting to maximise or minimise; it is set by how many conversions a week your budget needs to buy.
Running several ad sets against overlapping audiences adds a third fault: you bid against yourself, which Meta names as auction overlap.
5: The Click – Clicks but No Sales
Facebook ads getting clicks but no sales is the point where the ad has done its job. Someone saw it, believed it enough to click, and arrived. Whatever stopped them after that is on the page, not in Ads Manager.
Two things are worth confirming before you hand the problem over. That the ad and the page make the same promise, at the same price, for the same thing — a mismatch here reads as a lie and loses the visitor in seconds. And that the page works on a mid-range Android phone on mobile data, which is how most of this traffic arrives.
Beyond that, the diagnosis belongs elsewhere: conversion rate optimization — 10 reasons your visits don’t sell covers the page, and cart abandonment covers the final step for anyone selling online.
6: After the Lead – The Failure That Never Shows in Ads Manager
The last break is invisible to every report you run. The ad delivered, the click landed, the lead arrived — and then nobody answered the message for six hours, or the number was called once and never again, or the order was confirmed and the customer refused it at the door.
None of that appears in Ads Manager, because Ads Manager stops at the conversion event you defined. If the money is made on a phone call, in a shop, or in a conversation that ends days later, the ad account has no way to see it unless you send it back.
Meta supports exactly this. Offline conversions let you measure real-world outcomes such as in-store purchases, phone orders and bookings by uploading interaction data to a dataset in Events Manager, either manually or through the Conversions API. Once uploaded, those conversions are attributed to the people who saw or clicked your ads.
Two things follow for a business selling this way. Your reported cost per result is worse than your real one, because the wins are not being counted. And the delivery system is optimising on incomplete information, so it is finding more of the wrong people.
How to Know If It’s an Offer’s Problem?
You have an offer problem when every mechanical link checks out and the numbers still do not move. Delivery is stable, conversions are counted, the goal is right, the audience is the right size and fresh, the page works, and follow-up is fast. If all six hold and the result is still nothing, the ad is not the constraint.
Three signals point the same way. Click-through is reasonable but nothing converts, which means the promise is attractive and what follows it is not. Costs stay flat while creative changes, which means the message was never the variable. And the same offer performs no better through any other channel.
An offer problem is a price, a guarantee, a delivery time or a reason to choose you — not a headline. No amount of budget fixes it, and more spend just buys a faster answer to the same question.
What Conditions You May Not Be Able to Fix?
Some conditions sit outside the account entirely, and recognising them saves the budget you would spend fighting them. They fall into three groups: things the platform controls, things the market controls, and things your own operation controls.
Platform-side, ad review is not yours to hurry. Meta states that most ads are reviewed within 24 hours but that it can take longer, that enforcement is imperfect, and that ads may be re-reviewed after going live.
Market-side, the auction price is set by everyone else bidding for the same people. You can pay it, move to a different optimisation event, or wait out the season. You cannot lower it.
Operation-side, a stock-out, a delivery area you do not cover, or a price the market will not pay are all real limits. Ads reach people faster; they do not change what you are able to sell them.
When Does Fixing The Problem Just Not Worth it?
Fixing stops being worth it when the cost of the fix is larger than the value it can return. That comparison needs two numbers you should already have: what a customer is worth to you, and how far the current cost per result sits from that number.
A gap of a few percent is worth a diagnosis. A gap of several multiples usually is not, because closing it needs a different offer or a different channel, not a better campaign.
Two more cases end the argument. When the required budget cannot buy 50 conversions a week for even one ad set, the account cannot produce stable numbers at any level of effort, and the read is that this channel is not funded. And when the constraint is operational — the follow-up, the stock, the delivery — every pound spent on the ad account is spent on the wrong thing.
How Brand Brew Diagnoses a Campaign That Stopped Converting
BrandBrew runs the six checks in the order above rather than starting with creative, because creative is the most expensive thing to change and the least likely to be the fault.
The sequence is: confirm delivery and learning-phase status, reconcile counted conversions against actual orders, verify the optimisation event matches what the business sells, check audience size and saturation, test the landing experience on a mid-range Android device, then trace what happens to a lead after it arrives. The first break found is fixed and measured before anything below it is touched, because a fault higher in the chain makes everything below it unreadable.
Whichever link is broken, the account tells you — but only if the checks are run in an order where one answer does not corrupt the next.
Why are my Facebook ads active but not delivering?
Facebook ads can show as active and still not deliver because the ad set cannot spend at the pace you set. The usual causes are an audience too small for the optimisation event, a bid cap or cost per result goal below the current auction price, a budget too low to buy enough results, or heavy overlap between your own ad sets competing in the same auction.
What does “not delivering” mean on Facebook ads?
“Not delivering” is a Delivery column status meaning the ad is not currently being shown to anyone. It is a state, not a diagnosis. It covers ads still in review, ads rejected against Meta’s Advertising Standards, ad sets outside their scheduled dates, campaigns that have hit a spending limit, and ad sets whose settings make delivery impossible.
What is a good conversion rate for Facebook ads?
A good conversion rate for Facebook ads is any rate where your cost per conversion sits below what a customer is worth to you. Meta publishes no benchmark, and a cross-industry average would blend businesses with different prices, different optimisation events and different buying cycles. Your own rate from a stable period, compared against your margin, is the only usable standard.
Is a 12% conversion rate good?
A 12% conversion rate cannot be judged without knowing the denominator. Twelve percent of link clicks reaching a lead form is plausible and healthy. Twelve percent of impressions becoming purchases would be extraordinary and usually signals a measurement error or a mislabelled event. Confirm what the 12% is measured against before treating it as good or bad news.
How long should I wait before deciding my ads aren’t working?
Wait until the ad set has left the learning phase, which Meta says takes about 50 results in the week after the last significant edit. Before that point, performance is unstable and not indicative of future results. If your budget cannot buy 50 results in a week, the ad set will never reach a readable state, and waiting longer will not help.
Why did my ads work last month and stop this month?
Ads that worked last month and stopped this month usually met a more expensive auction rather than a broken campaign. More advertisers bidding for the same people means the same budget buys fewer impressions. The other common causes are creative fatigue and audience saturation, where the same people have seen the same ad too many times without acting.
Should I turn the ad off or let it run while it’s underperforming?
Leave it running if it has not yet left the learning phase, because pausing and restarting resets learning and repeats the unstable period you already paid for. Turn it off if it has stabilised and the cost per result sits clearly above what a customer is worth. Frequent pausing is named by Meta as something that disrupts delivery pacing.
Before You Spend Another Pound on Ads That Don’t Convert
The reason Facebook ads are not converting is almost never the reason people reach for first. Creative gets rewritten while the ad set is still in the learning phase. Budget gets raised while half the conversions are going uncounted. Audiences get rebuilt while the actual break is a lead nobody answered.
Run the six in order: delivery, measurement, optimisation, audience, the click, and what happens after the lead. Stop at the first break, fix it, and let the numbers settle before touching anything below it. Most accounts never need to reach step four.
If the six checks come back clean and the numbers still do not move, the constraint has moved out of the ad account and into the offer or the operation — and that is a different decision, made with the four numbers that decide where your budget goes rather than with a new set of creatives.
BrandBrew works with Egyptian businesses on this diagnosis: finding which of the six links is actually broken before more budget goes through it.
