In this guide
- How should you set up your Ads Manager columns?
- Which metrics actually matter, and what do they tell you?
- What does "results" really count?
- How do you know if your cost per result is good?
- How do you diagnose a rising cost per result?
- What do the relevance rankings mean?
- How should you use breakdowns?
- Which metrics can you safely ignore?
- What does a weekly reporting routine look like?
- Frequently asked questions
- Want someone to read the numbers with you?
Read a Meta Ads Manager report from the bottom of the funnel up. Start with results and cost per result, then check whether real customers followed. If cost per result rises, look upwards for the cause: conversion rate, link click-through rate, CPM (what you pay for reach) and frequency. Ignore likes and "CTR (all)".
Ads Manager shows dozens of columns, and most business owners end up staring at the wrong ones. This guide covers the handful that drive decisions and a routine for reading them.
How should you set up your Ads Manager columns?
Build one saved column preset with only the metrics you'll act on, so every report opens the same way. In Ads Manager, open the Columns menu, choose Customise columns, pick the metrics below, and save it as a preset.
A lean preset for lead generation and messaging campaigns:
- Delivery and attribution setting
- Amount spent
- Results and cost per result
- Messaging conversations started (if you run click-to-message)
- Leads (if you run lead forms or website leads)
- Reach, impressions and frequency
- CPM (cost per 1,000 impressions)
- Link clicks, CTR (link click-through rate) and CPC (cost per link click)
- Quality, engagement rate and conversion rate rankings
For e-commerce, add purchases, cost per purchase, purchase conversion value and ROAS (return on ad spend).
Which metrics actually matter, and what do they tell you?
Results tell you whether the campaign works; the supporting metrics tell you why.
| Metric | What it means | What to do with it |
|---|---|---|
| Results | Count of the outcome your campaign optimises for | Check it matches what you really want (messages, leads, purchases) |
| Cost per result | Amount spent ÷ results | Your headline efficiency number; compare against your target |
| Amount spent | Total spend in the period | Confirm pacing matches your budget |
| Reach | Unique people who saw the ad | Useful for audience size and saturation |
| Impressions | Total times the ad was shown | Divide by reach to understand frequency |
| Frequency | Average times each person saw the ad | Rising frequency with falling CTR signals fatigue |
| CPM | Cost per 1,000 impressions | Shows how expensive your audience and season are |
| CTR (link) | Share of impressions that led to a link click | Measures how compelling the ad is |
| CPC (link) | Cost per link click | Combines CPM and CTR into one cost |
| Conversion rate | Results ÷ link clicks (calculate it yourself if needed) | Tells you how well the page, form or chat flow converts |
| Relevance rankings | How your ad compares with competing ads for the same audience | Below average flags creative or offer problems |
What does "results" really count?
Results counts whatever the campaign was told to optimise for, so it changes meaning from one campaign to the next. A messaging campaign's result is a conversation started; a lead campaign's is a lead form submission or website lead; a sales campaign's is a purchase.
Two checks before trusting it:
- Is the result close to revenue? A conversation started isn't a booking. Match it against your CRM or inbox.
- What attribution setting is used? Meta's default is 7-day click and 1-day view, meaning a result is credited if someone clicked within seven days or viewed within one day. All the GRWW figures we quote use this default.
Reports with different attribution settings can't be compared directly, so keep it consistent.
How do you know if your cost per result is good?
Compare it with your own target, your history and what each customer is worth, not someone else's screenshot. Work out the most you can afford per lead: average sale value × share of leads that become customers × the share of revenue you're willing to spend on marketing.
For context, here's what we've seen across our anonymised accounts:
- Nine featured accounts: 6,600 enquiries, 1.46M combined reach and 2.47M impressions.
- Wedding photography (five accounts): 1,130 enquiries at an average of ₹58.55 each.
- A clothing brand: ₹5.43 average per conversation with a 2.97% CTR.
- An immigration consultancy: a messages ad brought 340 conversations at $0.11; the best lead form, 245 leads at $0.34.
These vary enormously by industry and market, which is exactly the point. Our guide to what a good cost per lead looks like on Meta goes into benchmarks more carefully.
How do you diagnose a rising cost per result?
Work backwards through the funnel; the first metric that worsened is usually the culprit. Cost per result is roughly CPM ÷ (CTR × conversion rate), so one of those three has moved.
- Did CPM rise? Costs often rise in competitive seasons such as festivals and end-of-year sales, or because your audience is small. Try broader targeting or more placements.
- Did CTR fall? Usually creative fatigue. Check frequency; if it's climbing, refresh the ads. Our creative testing framework helps you replace tired ads with proven ones.
- Did conversion rate fall? Something changed after the click: a slower page, a broken form, longer reply times, a price change.
- Did nothing change but results dropped? Check tracking. A broken Pixel or Conversions API connection can make a healthy campaign look dead.
If everything seems to have fallen apart at once, work through our checklist for when Facebook ads stop working.
Don't change anything until you know which number moved. Fixing creative won't help if the real problem is a broken form.
What do the relevance rankings mean?
Quality, engagement rate and conversion rate rankings compare your ad with others competing for the same audience. They're diagnostic, not a direct factor you can optimise.
- Low quality ranking: the ad may look like clickbait, use poor imagery or feel low quality. Improve the creative.
- Low engagement rate ranking: people aren't reacting. Try a stronger hook or format; Reels often help.
- Low conversion rate ranking: people engage but don't convert. Check the offer, the form or the page.
Rankings only appear once an ad has enough impressions, so don't read too much into new ads.
How should you use breakdowns?
Breakdowns split results by age, gender, placement, region, device or time, revealing where money works and where it leaks. Use the Breakdown menu in Ads Manager.
Useful checks:
- Placement: is one placement producing cheap results that never convert? Compare with CRM quality before excluding it.
- Region: are certain cities much cheaper or much better quality?
- Age: do leads from one age group close better?
- Time of day: helps decide reply-team schedules rather than ad scheduling.
Breakdowns can mislead on small numbers. Don't exclude a segment because of a handful of results.
Which metrics can you safely ignore?
Anything that doesn't connect to your goal: likes, post reactions, "CTR (all)" and reach on its own.
- CTR (all) counts any click, including clicks on your profile name or "see more". Link CTR is the honest version.
- Post engagement looks nice but doesn't pay bills unless you're running an engagement campaign deliberately.
- Reach matters for awareness campaigns; for lead generation it's context, not a goal.
- Video plays alone, unless you're building video-viewer audiences for retargeting.
What does a weekly reporting routine look like?
Ten minutes a week with the same preset and the same questions beats an hour of random clicking.
- Set the date range to the last seven days and compare it with the previous seven.
- Check spend, results and cost per result at campaign level.
- Look at frequency and CTR at ad level to spot fatigue.
- Compare results with your CRM or inbox: how many were genuine?
- Note one action for the week: refresh creative, adjust budget or fix a page.
For the bigger picture of tracking, landing pages and channel choices, see our performance marketing guide for 2026.
Frequently asked questions
What's the difference between reach and impressions?
Reach is the number of unique people who saw your ad. Impressions count every time it was shown, including repeats. Impressions divided by reach gives you frequency.
What is a good frequency for Meta ads?
There's no universal number. Watch the trend instead: if frequency rises while CTR falls and cost per result climbs, the audience is tiring of the ad and it's time for new creative.
Why don't Ads Manager results match my CRM?
Attribution windows, view-through credit, duplicate leads and tracking gaps all cause differences. Ads Manager credits results based on its attribution setting, while your CRM records what actually arrived. Use the CRM for business decisions.
Should I look at CPC or CPM?
Both, in context. CPM tells you how expensive it is to reach your audience; CPC adds how compelling your ad is. Cost per result matters most, and CPM and CPC help explain changes in it.
How often should I check my Meta ads report?
A quick daily glance for spend problems is fine, but make decisions weekly. Daily numbers swing too much, especially on smaller budgets, and reacting to them often resets learning.
Want someone to read the numbers with you?
We'll walk through your Ads Manager account, set up a reporting preset and connect results to real enquiries. See how we work with clients or book a reporting review.