A facebook ads report can show dozens of columns at once: reach, impressions, frequency, click-through rate, cost per result, ROAS, engagement, video views, and more. Most of that is noise for a small business trying to answer one question: is this money being spent well. Reading the report well means knowing which numbers actually answer that question and which ones sound important but rarely do.
This guide covers eight numbers genuinely worth watching, four that commonly mislead people, and how to check the whole report against reality rather than trusting it blindly.
Setting up a facebook ads report that is honest
Before looking at any number, get the basics right. Make sure the pixel or Conversions API is tracking correctly, confirm the attribution window is one you understand, and pick a small set of columns to look at every time rather than scanning the full default table. A report built on shaky tracking will mislead you no matter how carefully you read it, so this step is worth doing once, properly, before trusting anything downstream.
The numbers that matter: results and cost per result
Whatever you optimised the campaign for, purchases, leads, add-to-carts, this is the number that tells you whether the campaign is doing its job, and cost per result tells you what each one is costing. Together they are the two most honest numbers in the account, because they are tied directly to the objective you actually set.
Watch the trend over at least a week rather than a single day. Daily numbers on a modest budget bounce around for reasons that have nothing to do with campaign quality.
The numbers that matter: spend and pacing
Knowing exactly what has been spent, and whether it matches the budget you set, catches problems early: a campaign left running after it should have paused, or one barely spending because of a delivery issue. This is a simple number, but it is worth checking on its own rather than only inferring it from the total at the bottom of the report.
The numbers that matter: frequency
Frequency shows how many times, on average, the same person has seen the ad. Rising frequency alongside falling results is one of the clearer early signs that an audience is becoming exhausted with the creative, which is worth catching before cost per result climbs noticeably. On its own, frequency means little; paired with the results trend, it is genuinely useful.
The numbers that matter: link click-through rate and conversion rate on site
Link CTR shows whether the ad itself is compelling enough to earn a click. Your website's own conversion rate for that traffic shows whether the landing page then does its job. Reading these together separates an ad problem from a page problem: strong CTR with a weak on-site conversion rate usually means the page, not the ad, needs attention, and the reverse points the other way.
The numbers that matter: ROAS, read carefully
Return on ad spend is useful once you understand what it does and does not include. It typically reflects revenue reported by the ad platform against spend, not your actual profit after product cost, shipping, returns and payment fees. Treat it as a directional signal for comparing campaigns against each other, not as a substitute for checking real profitability in your own accounting.
The numbers that mislead: reach on its own
Reach tells you how many people saw the ad at least once, and on its own it says nothing about whether any of them cared. A campaign can post an impressive reach number and produce almost no results. Treat reach as context for interpreting other numbers, never as a success metric by itself.
The numbers that mislead: CPM out of context
Cost per thousand impressions moves for reasons that have nothing to do with your campaign's quality, including seasonal competition for ad space. A rising CPM alone does not mean a campaign is performing worse; check it alongside cost per result before drawing any conclusion.
The numbers that mislead: likes, comments and shares
Engagement can feel like proof an ad is working, and it does help an ad's relevance to some degree, but it does not reliably predict sales or leads. A widely liked ad selling nothing is not a marketing success. Track engagement as one input among several, not as a goal in itself unless brand awareness is genuinely the objective you set.
The numbers that mislead: "results" against the wrong objective
If a campaign was set to optimise for landing page views but the report is read as if it optimised for purchases, the reported "results" measure the wrong thing entirely and will look misleadingly strong or weak. Always check what a campaign was actually optimising for before judging its results column.
Attribution settings change the story
The attribution window you use, meaning how long after seeing or clicking an ad a purchase still gets credited to it, changes how many results the same campaign appears to produce. A longer window generally shows more results for the same spend, because it credits the ad for more delayed decisions. Neither a short nor a long window is objectively correct; the important thing is consistency, so that comparisons between weeks or campaigns are comparing like with like rather than a moving target. Meta's own explanation of how attribution works is worth reading once, because it changes how every other number in a facebook ads report should be interpreted.
Relevance diagnostics, read correctly
Alongside the main results table, Meta also surfaces relevance-related diagnostics that compare an ad's quality, engagement and conversion signals against similar ads competing for the same audience. These are relative rankings, not absolute scores, so a "below average" reading means the ad is underperforming similar competitors for that audience, not that it has failed some fixed standard. They are a useful early warning when cost per result is climbing but the cause is not yet obvious, and Meta's own guide to ad relevance diagnostics explains what each one actually measures.
Checking a facebook ads report against your own sales data
The platform's numbers are estimates, not a ledger. Periodically compare what Ads Manager reports against what actually happened in your store or CRM over the same period. A small, steady gap is normal and comes from attribution differences. A large or growing gap is worth investigating, since it often points to a tracking issue such as duplicated events, a broken pixel on part of the site, or purchases happening through a channel the tracking does not see.
Reading a report week to week, not day to day
Daily fluctuations on a modest budget are usually just statistical noise rather than a signal. Reviewing performance on a weekly rhythm, while keeping a light daily glance for anything obviously broken such as a paused payment method or a rejected ad, avoids reacting to noise while still catching real problems quickly. Constant daily changes based on small day-to-day swings tend to do more harm than good, because they interrupt the learning that stabilises results over time.
Building a simple column preset
Ads Manager lets you save a custom set of columns, and it is worth setting one up with only the numbers covered above: results, cost per result, spend, frequency, link CTR, on-site conversion rate, and ROAS as a directional check. Removing the rest from view reduces the temptation to react to numbers that were never meant to drive decisions on their own.
When to worry and when to wait
Worth investigating promptly: cost per result rising steadily over several days rather than one, frequency climbing alongside falling CTR, or a growing gap between the platform's reported results and your own sales. Worth waiting out: a single unusual day, small week-to-week swings on a modest budget, or a CPM increase with cost per result still stable. Learning to tell these apart is most of what makes a facebook ads report genuinely useful rather than a source of constant, unproductive worry.
A short weekly routine for reading a facebook ads report
A simple routine keeps this manageable. Once a week, open the saved column preset, compare cost per result and frequency against the previous week, and glance at whether spend paced roughly as planned. Once a month, check the platform's reported results against your store or CRM figures for the same period, so drift in tracking gets caught before it becomes a large, confusing gap. This takes a few minutes and catches the great majority of problems worth catching, without turning ad management into a daily source of stress over numbers that were never meant to be read that closely.
New advertisers often try to read every column every day. Experienced ones read a short weekly report closely and skim the rest. That shift alone tends to improve decision-making more than any single metric on the page, because it stops small, meaningless swings from triggering changes that interrupt a campaign's learning.
What a healthy trend looks like over a month
Individual weeks are noisy, so it helps to know what a healthy pattern looks like over a longer stretch. Cost per result that moves within a reasonably narrow band from week to week, without a sustained upward slope, is generally a sign that a campaign has settled rather than deteriorated. A steady or slowly improving trend over a month is a far more reliable signal than any single week, good or bad, and it is the pattern worth showing anyone outside the day-to-day management of the account.
Equally, a single strong week following months of steady performance is not necessarily proof that something new is working; it may simply be normal variation. Waiting for a pattern to hold across at least a few weeks before crediting it to a specific change avoids drawing conclusions from what is really just noise in a facebook ads report.
Sharing the report with someone else
If someone other than the person running the campaigns reviews the report, whether a business partner, an accountant or an agency client, give them the same short column preset rather than the full default view. A facebook ads report full of unfamiliar columns invites the wrong questions, usually about numbers like reach or engagement that were never meant to carry the weight of "is this working". A short, consistent report, paired with a plain-language note on what changed and why, keeps everyone focused on the numbers that actually answer that question.
Getting more from your reporting
A report is only as good as the tracking behind it and the discipline applied to reading it. If your numbers feel unreliable, or you are not sure which of them to trust, that is usually a tracking or attribution question worth fixing before changing anything in the campaigns themselves. Our Facebook and Instagram ads management service includes setting up reporting you can actually trust. See our note on building tracking you can trust for the setup side of this, or get in touch if your reports are not adding up.