Short answer: The Facebook ads metrics that matter most are the ones tied to money: cost per acquisition (CPA) and return on ad spend (ROAS), checked against your real sales or CRM. Below those sit diagnostic metrics such as CPM, link click-through rate, frequency and hook rate, which explain why CPA or ROAS moved. A good Meta Ads specialist reads them in that order, top down, and never optimises a diagnostic metric at the expense of the business result.
Facebook ads metrics: think in layers
Ads Manager offers hundreds of columns. Most of them are noise for day-to-day decisions. I find it helps to group the useful ones into three layers:
- Business results: CPA, ROAS, conversion volume and, outside Meta, revenue, profit and new customers.
- Traffic and auction: CPM, link click-through rate, cost per link click and frequency.
- Creative attention: hook rate, hold rate and how many people actually watch or engage.
Layer one tells you whether things are working. Layers two and three tell you where to look when they are not. If CPA rose, was it because ads got more expensive to show (CPM), fewer people clicked (CTR), or fewer clickers bought (conversion rate on the site)? Each answer leads to a different fix.
Layer 1: CPA, ROAS and conversions
CPA (shown as cost per result in Ads Manager) is spend divided by the number of conversions. ROAS is conversion value divided by spend. Lead generation accounts focus on CPA, ideally cost per qualified lead. eCommerce accounts usually watch both.
Three rules I follow with these numbers:
- Know your target before you look. A good CPA is one below the most you can afford to pay for a customer, based on your margins. A number on its own means nothing.
- Compare with your back end. Check Meta’s reported purchases or leads against your store or CRM every week. Gaps happen, and you need to know their size.
- Judge over enough time. Daily CPA swings are normal. Make decisions on weekly or longer windows, unless something is clearly broken.
Attribution: why your numbers changed in 2026
CPA and ROAS depend on how Meta credits conversions to ads, and that changed twice this year. In January 2026, Meta removed the 7-day view and 28-day view attribution windows from the Ads Insights API, leaving 1-day view alongside the click windows. Then, from March 2026, click-through attribution started requiring an actual link click. Likes, comments, saves and shares now sit in a separate engage-through category with a 1-day window.
What this means in practice:
- Reported conversions can drop even when real sales did not. Compare against your own sales data before reacting.
- Do not compare this year’s figures directly with older periods without noting the change. Set a new baseline.
- Use the “compare attribution settings” option in Ads Manager to see how much of your result comes from clicks versus views and engagement.
Layer 2: CPM, CTR, cost per click and frequency
| Metric | What it measures | What a change usually means |
|---|---|---|
| CPM | Cost per 1,000 impressions | Auction pressure, audience size, seasonality or low ad quality |
| Link CTR | Link clicks divided by impressions | Whether the ad persuades people to click |
| Cost per link click | Spend divided by link clicks | A mix of CPM and CTR |
| Frequency | Average times each person saw the ad | Audience saturation and creative fatigue |
| Landing page views | Clicks that loaded your page | Page speed and accidental clicks |
A few notes from how I read these:
- Use link CTR, not CTR (all). CTR (all) counts any click, including profile taps and “see more”. Link CTR is the one connected to traffic.
- A high CPM is not always bad. Reaching buyers in competitive markets costs more. If CPA is on target, a high CPM is fine.
- Watch the gap between link clicks and landing page views. A big gap often points to a slow page or poor mobile experience, which my conversion rate optimisation work usually starts with.
- Read frequency as a trend. There is no universal safe number. Rising frequency with falling CTR and rising CPA is the warning sign, especially in small warm audiences like the ones in Meta Ads retargeting.
Layer 3: hook rate and hold rate for video
With broad targeting doing most of the audience work, creative is the biggest lever you control. Two custom metrics help judge video creative early:
- Hook rate = 3-second video plays divided by impressions. It shows how well the first seconds stop the scroll.
- Hold rate = ThruPlays divided by 3-second video plays. It shows whether people stay once they have stopped.
Neither is a default column. As Vaizle’s guide to hook rate and hold rate shows, you create them as custom metrics in Ads Manager using the formulas above. I use them to diagnose, not to decide. A video with a high hook rate and poor CPA may be attracting the wrong attention. A video with a modest hook rate and a strong CPA is a winner. Business results always overrule attention metrics.
How I diagnose a rising CPA
When cost per result climbs, I work through the layers in this order:
- Check tracking. Did conversions drop in Meta but not in the store or CRM? Then it is a measurement issue, not a performance one. My tracking and analytics service covers this check.
- Check recent changes. Budget jumps, new ad sets or big edits can push campaigns back into learning.
- Check CPM. If CPM rose sharply with no change in CTR, look at seasonality, competition and audience size.
- Check link CTR and frequency. Falling CTR with rising frequency usually means creative fatigue. Time for new ads.
- Check the site. If CTR is steady but conversions fell, look at the landing page, prices, stock and checkout.
- Check the audience mix. Audience segment breakdowns show whether spend shifted between new people and existing customers. My guide to Facebook ads targeting explains why that split matters.
A simple weekly metrics view
What I recommend is one saved column set in Ads Manager with: amount spent, results, cost per result, purchase ROAS (for eCommerce), CPM, link CTR, cost per link click, landing page views, frequency, hook rate and hold rate. Review it weekly at campaign and ad level. Then put the Meta numbers next to your real revenue, leads and new customers in a simple sheet, so you can see whether the platform story matches the business story.
That second step is where most reporting goes wrong. Meta’s numbers are useful for steering campaigns, but your bank account is the final judge.
Frequently asked questions
Which Facebook ads metrics matter most?
What is a good CTR for Facebook ads?
Why did my Meta conversions drop when sales did not?
How do I add hook rate in Ads Manager?
Who can review my Facebook ads metrics and reporting?
Need help reading your Meta Ads numbers?
If your reports are full of numbers but you still cannot tell whether ads are making money, I can help. I work as a Meta Ads specialist for brands in the US, UK, Canada and Australia and set up reporting that ties ad spend to real results. Start with a free ads audit and I will show you which metrics to watch.