Most advertisers answer this question by looking at ROAS. If ROAS is up, the ads are working. If ROAS is down, something is wrong.
That’s understandable — ROAS is the number that connects directly to revenue. But it’s also the worst metric to use as your primary diagnostic tool, because it tells you the outcome without telling you anything about the cause. A ROAS of 2.5x this week versus 3.8x last week tells you results have dropped. It tells you nothing about whether the drop is coming from your creative, your audience, your landing page, your offer, or something external to your campaigns entirely.
Reading four metrics together, rather than ROAS in isolation, is what gives you an actual picture of what your campaigns are doing — and more importantly, why.
The Four Metrics That Tell You What’s Really Happening
CTR — is your ad earning attention?
CTR, or click-through rate, measures the percentage of people who see your ad and click it. It is the most direct signal of whether your creative is working — whether the hook is landing, whether the visual is stopping the scroll, whether the offer is compelling enough to prompt action at first exposure.
A commonly used reference point for link CTR on cold traffic is around 1%. Below that is worth examining. Above that generally indicates the creative is resonating. But the absolute number matters less than the trend — a CTR that has moved from 1.8% to 0.9% over three weeks is telling you something regardless of whether 0.9% looks acceptable in isolation.
What a low or declining CTR tells you: the creative is the problem. Either the ad is fatiguing with an audience that’s seen it too many times, or the hook isn’t connecting with the audience you’re targeting. The fix is in the ad itself — copy, creative, angle, or format — before anything else.
What a healthy CTR tells you: the ad is doing its job. If results are still poor despite a strong CTR, the problem is not the ad.
CPM — are you reaching your audience efficiently?
CPM, or cost per thousand impressions, measures how expensive it is to reach your audience. It is partly in your control (through audience selection and creative quality) and partly driven by external factors (auction competitiveness, seasonality, platform changes).
Rising CPM means you are paying more to reach the same number of people, which compresses ROAS independently of how well the ad is performing. Falling CPM means you are becoming more efficient — often a signal that the algorithm has found a better-performing audience within your parameters.
What high or rising CPM tells you: your targeting may need adjustment, the audience may be saturated, or external competitive pressure has increased. Broad audiences and advantage+ setups sometimes deliver lower CPM than tightly defined interests — worth testing if CPM is consistently high.
What CPM alone cannot tell you: whether the audience you’re reaching at that CPM is the right one. A low CPM reaching the wrong people is still a waste. CPM needs to be read alongside CTR and conversion data to be meaningful.
Frequency — are you over or underreaching your audience?
Frequency measures how many times the average person in your audience has seen your ad. It is one of the most underread metrics in Meta advertising, and one of the most informative.
Below 2 for cold traffic: most people in your audience haven’t seen the ad enough for it to register. Low frequency on a prospecting campaign isn’t necessarily a problem if reach is growing — it means you’re finding new people. But very low frequency alongside high spend can indicate inefficient delivery.
Between 2 and 4 for cold traffic: a healthy range for most prospecting campaigns. The audience is seeing the ad without being saturated by it.
Above 4 for cold traffic: worth watching. You’re starting to show the same ad to the same people repeatedly, which typically produces diminishing engagement. CTR usually starts to fall at this point.
Retargeting campaigns operate on different frequency norms — higher repetition is expected and often appropriate since you’re reinforcing intent rather than generating initial awareness. The 2 to 4 benchmark applies specifically to prospecting.
The action funnel — where does interest become purchase?
The action funnel is the sequence between someone clicking your ad and completing a purchase. It’s where the majority of conversion problems live — and the metric that most advertisers least regularly look at.
A simplified version of the funnel: click → landing page view → add to cart → initiate checkout → purchase.
Each step in this sequence has a drop-off rate. Some drop-off is normal. But a large drop-off at a specific step is a specific signal:
Heavy drop-off between click and landing page view: the page is slow to load or there is a significant mismatch between what the ad promised and what the page delivers. The user bounced before seeing the offer.
Heavy drop-off between landing page and add to cart: the offer is not landing on the page itself — pricing, copy, trust signals, product presentation. The user arrived interested and left unconvinced.
Heavy drop-off between add to cart and checkout: friction at the checkout stage — too many steps, unexpected costs, a required account creation, or a payment method not available.
Heavy drop-off between checkout initiation and purchase: the checkout itself has a usability problem, or there is a trust issue at the payment stage — security signals, unfamiliar payment gateway, or doubts about the brand.
Identifying which step has the largest drop-off tells you exactly where to invest attention. Fixing the ad delivers nothing if the problem is a slow landing page. Fixing the checkout delivers nothing if the problem is the landing page copy.
Reading the Four Together
The real value of this framework is not in each metric individually — it is in what they tell you in combination.
High CTR + high CPM + low conversion: The ad is compelling and the audience is interested, but it is expensive to reach them and something is breaking down after the click. Check the funnel.
Low CTR + rising frequency: The creative has fatigued. The same audience has seen the ad too many times. Refresh the creative before adjusting anything structural.
Stable CTR + stable CPM + falling ROAS: The ad and audience are consistent. The drop is happening downstream — look at your funnel drop-offs and check whether anything changed on the landing page or in the offer.
Low CTR + low frequency + low CPM: The campaign may be in early learning phase or targeting a genuinely cold, broad audience that needs more time to warm up. Avoid making structural changes before the algorithm has had enough events to optimise.
All metrics stable, ROAS volatile week to week: Normal fluctuation. A single week’s variance is almost never a signal worth acting on. Look at the 3 to 4 week trend before drawing conclusions.
What This Framework Cannot Tell You
Reading these four metrics gives you a clearer picture than most advertisers have of what their campaigns are doing. It does not tell you everything.
It doesn’t tell you whether your offer is fundamentally positioned correctly for your market. It doesn’t tell you whether your pricing is competitive. It doesn’t tell you how your brand is perceived relative to alternatives. Those are bigger questions that live upstream of any campaign — and no amount of Meta ads optimisation fixes a problem that exists at that level.
The four-metric diagnostic is built for the scenario where the business and offer are sound, and the question is why the campaigns are not reflecting that. If the offer itself is the issue, that conversation is different.
The Difference Between Working and Optimised
One distinction worth being clear about: working and optimised are not the same thing.
A campaign can be working — generating profitable ROAS, spending consistently, producing results — while still having significant room for improvement. CTR could be lifted with better creative. CPM could be reduced with audience adjustments. Funnel drop-offs could be closed with page changes. The existence of results does not mean the campaigns are running at their ceiling.
The four-metric framework is useful both for diagnosing what’s broken and for identifying where a functioning campaign can be made significantly more efficient. The difference between a 2x and a 3.5x ROAS is often not a structural campaign change — it is a series of incremental improvements across the funnel that compound over time.
This breakdown is written by Jason Gan, a Meta Certified Professional and Badged Meta Business Partner who has personally audited over 1,000 advertiser accounts since 2010. You can see real account breakdowns on the Jason Gan YouTube channel.
No login required. Just a screen share, your real numbers, and a plan you can act on the same day.
If you want to build the ability to run this diagnostic yourself week by week, 1-on-1 coaching is built around exactly this framework — applied to your account, your data, your specific campaigns.
Frequently Asked Questions
How do I know if my Meta ads are working?
Read four metrics together rather than ROAS in isolation: CTR (is the creative earning attention), CPM (are you reaching your audience efficiently), frequency (are you over or underreaching your audience), and the action funnel (where between click and purchase are people dropping off). ROAS tells you the outcome — these four tell you why you’re getting it.
What is a good ROAS for Meta ads?
There is no universal good ROAS — it depends entirely on your margins, average order value, and business model. The more useful question is what ROAS you need to be profitable given your specific cost structure, and working backward from that number. A 2x ROAS might be highly profitable for one business and deeply unprofitable for another.
Why are my Meta ads getting clicks but not converting?
Almost always a problem downstream of the ad — the landing page, offer, or checkout. The ad is working. Something after the click is losing the interest it generated. Diagnose which step in the funnel has the largest drop-off, and fix that step rather than the ad.
What is a good CTR for Meta ads?
Around 1% link CTR is a commonly used reference point for cold traffic. The direction of movement matters more than the absolute number — a CTR declining week over week is worth investigating regardless of where it sits relative to a benchmark.
Part of the Meta Ads Coaching resource library — explore all articles by topic.