2026-09-20 · 6 min read · MetriQuill Team

Meta Ads Metrics to Include in Client Reports

Choose report metrics by campaign objective, not by whatever Ads Manager happens to show. Sales campaigns are judged on purchases, cost per purchase and ROAS. Lead campaigns are judged on leads and cost per lead. Traffic campaigns are judged on link clicks and cost per link click. Awareness campaigns are judged on reach, frequency and CPM. Aim for five to seven headline metrics per objective and move everything else to an appendix.

The rest of this guide gives you the table, the formulas, and a way to read metrics in pairs so the report explains why numbers moved instead of just listing them.

Metrics by campaign objective

Meta groups campaigns under a handful of objectives. These four cover most agency work:

Objective	Headline KPIs	Supporting metrics	Diagnostic metrics	Usually leave out
Sales	Purchases, cost per purchase, ROAS	Amount spent, link click-through rate, average order value from the store	CPM, frequency, landing page views	Reach, likes, post engagement
Leads	Leads, cost per lead	Amount spent, link click-through rate, conversion rate	CPM, frequency, lead quality from the client's CRM	ROAS (unless leads carry a value), reach
Traffic	Link clicks or landing page views, cost per link click	Amount spent, link click-through rate	CPM, frequency, share of clicks that become landing page views	Purchases, ROAS
Awareness	Reach, frequency, CPM	Amount spent, video views if the ads are video	Click-through rate as a rough attention signal	Cost per purchase, ROAS

A quick test for any metric: if this number doubled or halved, what would we do differently? If you cannot name an action, it belongs in the appendix.

The core metrics, defined

Metric	Formula	How to read it
CTR (link click-through rate)	Link clicks / impressions x 100	Whether the creative and audience earn the click. Compare within an objective, not across them
CPC (cost per link click)	Amount spent / link clicks	The cost of traffic. A cheap click that never converts is not cheap
CPM	Amount spent / impressions x 1,000	The cost of attention. It rises with competition and seasonal demand
Cost per result	Amount spent / results	Only comparable between campaigns with the same result type. See Results explained
Conversion rate	Results / link clicks x 100	What happens after the click. It depends on the page and the offer more than on the ad
ROAS	Purchase value / amount spent	Revenue per unit of spend. Whether it is good depends on margin, not on a universal number
Frequency	Impressions / reach	Average number of times each person saw the ads. Watch the trend
Reach and impressions	Unique people vs total views	Reach counts people, impressions count views. Frequency connects them

One detail that trips people up: "clicks (all)" includes reactions, comments and shares, while link clicks count only clicks that lead to the destination. For traffic and conversion reporting, use link clicks and the link click-through rate.

Vanity metrics versus decision metrics

Reach, likes and comments are not useless. They are useful in an awareness campaign, where the goal is attention. They are noise in a sales report, where the client is judged on revenue. When you hand a sales client a page of reach and engagement, you are giving them numbers that cannot lose, and they will notice.

Keep vanity metrics for context, keep decision metrics for the headline.

Read metrics in pairs

A single metric rarely explains anything. These pairs point at causes. Treat them as rules of thumb, not laws: each one gives you a place to look next, not a verdict.

You see	It often means	Check next
High CTR, low conversion rate	The ad promises something the page or offer does not deliver	Landing page match, page speed, offer, conversion tracking
Low CTR, normal CPM	The creative or audience is not earning attention	New hooks and creative, audience overlap
CPM jumps, CTR flat	More competition or seasonal auction pressure	Compare with the same period last year, placement mix
Frequency rising while CTR falls	Audience fatigue	Refresh creative, widen the audience
Spend up, results flat	Efficiency dropped, or measurement changed	Budget changes in the period, learning phase, attribution changes
Cost per result down, volume down	Delivery narrowed to cheaper but smaller pockets	Budget and bid limits, audience size

The last two rows deserve special care in 2026. Meta changed how it attributes conversions in January and March, so flat or falling results can be a measurement effect. See Meta attribution changes 2026.

How many metrics is enough?

Five to seven per objective. More than that and the reader stops looking for the story. If a client asks for a metric you left out, add it to the appendix rather than the headline.

Accounts with mixed objectives

If a client runs lead, traffic and sales campaigns together, give each objective its own section and its own headline KPIs. Do not add up "Results" across them, because a total of leads, clicks and purchases is a number with no meaning.

Grade against benchmarks the right way

A metric only becomes a judgment when you compare it with something. Compare within the same objective, against the client's own trailing history first and against published medians second. Benchmarks are guardrails, not targets. Current medians and how to use them are in Facebook Ads benchmarks 2026.

MetriQuill applies the same logic automatically: it infers each campaign's type from words in the campaign name (lead, traffic, conversion, awareness, retargeting) and grades it against benchmarks for that type, using published medians from WordStream and Triple Whale where they exist. Campaigns with no recognizable word fall back to general defaults, so naming campaigns with their objective pays off. You can see how it looks by uploading an export to the free tool.

Frequently asked questions

What is a good CTR on Facebook ads? It depends on the objective. WordStream's 2026 medians are about 1.9% for traffic campaigns and 2.7% for lead campaigns in the US, but your own history matters more. Details are in the benchmarks guide.

Should I report ROAS for a lead-generation client? Only if each lead carries a value. Otherwise report cost per lead and, ideally, lead quality from the client's own CRM.

What frequency is too high? There is no universal number. The signal is frequency rising while click-through rate falls. As rules of thumb, MetriQuill warns above 2.5 and treats above 4 as high, and retargeting audiences naturally run higher because they are smaller.

Reach or impressions? Show both only when they answer a question, and let frequency tie them together. For most sales and lead clients, neither is a headline metric.

Which metrics should I drop first? Anything you cannot connect to an action: reach for a sales client, likes for a lead-generation client, and any number the client has never asked about.

Last reviewed September 2026.