Ads Manager shows dozens of numbers, and most of them don't tell you whether your ads are making money. This plain-words guide explains reach, impressions, CPM, CTR, CPC, cost per result and ROAS, which ones to check every week, and how to tell when a good-looking number is hiding a problem.
Open Facebook Ads Manager or Google Ads for the first time and you'll see a wall of numbers: reach, impressions, frequency, CPM, CTR, CPC, results, cost per result, ROAS. Each one sounds important. Some agencies send reports filled with them, and it's easy to come away thinking things are going well when they aren't, or badly when they're fine.
You don't need to understand every column. You need to know which number tells you whether ads are making money, and which numbers help explain why. This guide covers both, in plain words, with simple examples in rupees.
The one number to look at first
Every campaign has a goal: a sale, an enquiry, a booked call, a WhatsApp message. The cost per result is how much you paid in ads for each of those. In Ads Manager, the result is whatever goal you set for the campaign, so make sure it's set to something that matters to your business.
Then compare it with what a customer is worth to you. If an online store makes ₹600 profit on an average order and pays ₹350 in ads for each sale, the ads are working. If a coaching institute earns ₹15,000 from a student and pays ₹800 per enquiry, it depends on how many enquiries become students. If one in ten does, each student costs ₹8,000 in ads, which is healthy. If one in fifty does, each costs ₹40,000, which isn't.
That's why our guide to Facebook ad budgets starts with what a customer is worth. Every other number in your report only makes sense next to it.
ROAS: return on ad spend
For online stores, the report also shows ROAS, short for return on ad spend. It's the sales value credited to your ads divided by what you spent. Spend ₹10,000, get ₹40,000 in sales, and your ROAS is 4 (often written as 4x).
ROAS is useful, but it measures revenue, not profit. A 3x ROAS is excellent for a brand with 70% margins and a loss for one with 25% margins. Work out your break-even ROAS, the return at which ads pay for themselves:
- Break-even ROAS = 1 ÷ your profit margin (before ad costs).
- With a 50% margin, break-even ROAS is 2. Anything above 2 makes money.
- With a 25% margin, break-even ROAS is 4. A 3x ROAS loses money on every sale.
Remember that the first order isn't always the whole story. If customers come back and buy again, a first sale at break-even can still be very profitable over a year.

The numbers that explain why
When cost per result goes up or down, these numbers help you find the reason. Think of them as a chain: people see your ad, some click, some of those take action.
Reach, impressions and frequency
Reach is how many different people saw your ad. Impressions is how many times it was shown in total. Frequency is impressions divided by reach, the average number of times each person saw it. If 10,000 people saw your ad 25,000 times, the frequency is 2.5.
Frequency matters most. When it climbs high over a short period, the same people are seeing the same ad repeatedly, and results usually drop. That's a sign you need new ads or a wider audience.
CPM: the cost of being seen
CPM is the cost per 1,000 impressions. It's the price you pay for attention, and it's set by an auction against other advertisers. CPM usually rises in busy seasons like Diwali and other festival sales, when many brands compete for the same people.
You don't control CPM directly, so don't panic about it on its own. If CPM goes up but sales keep pace, you're fine. If it goes up and cost per result goes up with it, consider whether to ride out the busy period or adjust your budget.
CTR: are people interested?
CTR, click-through rate, is the percentage of people who clicked after seeing your ad. 50 clicks from 5,000 impressions is a 1% CTR. A low CTR usually means the ad isn't grabbing attention or the offer isn't appealing to the people seeing it.
In Meta, look at link click-through rate, not plain CTR. Plain CTR counts any click, including clicks on your page name, "see more" or the photo. Link CTR counts only clicks that go to your website or form, which are the ones that matter.
CPC: the cost of a visit
CPC, cost per click, is what you paid for each click. It depends on both CPM and CTR: a more interesting ad gets more clicks from the same impressions, so each click costs less. Again, use the link version (cost per link click) in Meta.
Conversion rate: what happens after the click
Once people land on your website, what share of them buy or enquire? That's your conversion rate. If your CTR is healthy and your CPC is reasonable but sales are low, the problem is probably not the ad. It's the page: slow loading, unclear prices, a long form, missing reviews or a checkout that asks for too much.
Putting it together: a quick diagnosis
When cost per result rises, walk down the chain to find where things changed:
- Did CPM rise? Competition or season. Hold steady or adjust budget, and refresh ads.
- Did link CTR fall? People are tired of the ad or it isn't appealing. Check frequency and try new creative.
- Did the website conversion rate fall? Something changed on the page: price, stock, speed, a broken form. Check the site yourself on a phone.
- Has nothing changed except cost per result? Check that tracking still works in Events Manager. A broken pixel looks exactly like a sudden collapse in results.

Numbers that look good but don't pay
Some numbers are easy to grow and feel like progress, but don't tell you much about sales:
- Likes, comments and shares on ads. Nice to have, but people who like an ad aren't necessarily people who buy.
- Video views, especially 3-second views. Almost anyone scrolling past counts.
- Page likes and followers from ads. Useful for some goals, but not a sign that ads are profitable.
- Reach on its own. Reaching 1 lakh people sounds great until you see that none of them bought.
If a report mostly talks about these and rarely mentions cost per sale or enquiry, ask why.
Why Ads Manager and your sales don't always match
You'll often find that Ads Manager says 40 sales came from ads while your store shows 30 orders that week, or the other way round. Both can be right, because they count differently:
- Meta's default attribution credits a sale to an ad if someone bought within 7 days of clicking it or within 1 day of seeing it. A person who saw your ad, then searched Google for your brand and bought, may be counted by both Meta and Google.
- Some sales from ads aren't tracked at all, because of browser privacy settings, people switching devices or buying on WhatsApp.
- Sales are reported on the day of the ad interaction, not always on the day of the order.
Treat Ads Manager as a guide for comparing ads with each other, and your actual sales, orders and enquiries as the final word. If total sales grow as ad spend grows, and cost per customer stays below what a customer is worth, the ads are doing their job.
A simple weekly check
You don't need to look at your ads every day. Once a week, ten minutes is enough:
- Total spend this week, and total sales or enquiries from your own records.
- Cost per result in Ads Manager, compared with last week and with what a customer is worth.
- Which ads brought the cheapest results. Keep those, pause the clear losers.
- Frequency on your main ad sets. If it's climbing, plan new ads.
- A quick visit to your own website on your phone, to check everything still works.
Numbers to ignore, and numbers to never ignore
- Mostly ignore: likes, post reactions, reach on its own, video views without a link to results.
- Use to diagnose: CTR, CPM, hook rate, frequency.
- Never ignore: spend, cost per result, ROAS against break-even, and the real number of customers in your store or CRM.
A 10-minute weekly report routine
- Write down last week's spend and results next to the week before.
- Check cost per result against your target.
- Look at which ads got the most spend and whether they deserved it.
- Check frequency on your main ad set for signs of fatigue.
- Compare platform results with real orders or leads.
- Note one decision: what to add, pause or change next week.
If results suddenly drop, the Andromeda audit walks through the usual causes in order.
A worked report, line by line
An illustrative month for an online store: ₹60,000 spent, 4,80,000 impressions, a CPM of ₹125, 6,000 link clicks, a CTR of 1.25%, 240 purchases, a cost per purchase of ₹250, and ₹3,60,000 in purchase value, so a ROAS of 6.
- What matters most: ₹250 per purchase and ROAS 6, checked against the store's break-even ROAS of, say, 3.
- What explains it: CTR shows the ads interest people; the 4% click-to-purchase rate shows the store converts.
- What to check against reality: of the 240 purchases, how many were delivered? With 15% COD refusals, it's about 204 delivered orders, so real cost per delivered order is about ₹294.
- What to watch next month: frequency and CPM, especially before festivals.
Why Meta, Google and your store disagree
Meta counts a purchase within its attribution window, Google counts its own, and both may claim the same sale. Your store counts each order once. Use each platform's numbers to compare its own campaigns, and judge the whole on total store sales against total ad spend. Our ad tracking guide explains the setup behind the numbers.
Quick quiz · 4 questions
Read the report: quiz
1. You spent ₹50,000 and sales from ads were ₹2,00,000. What's the ROAS?
2. Your break-even ROAS is 3.5 and your ROAS is 3. Are ads profitable?
3. CPM rose sharply in late October. The most likely reason?
4. Which number should decide if a campaign works?
0 of 4 answered
Frequently asked questions
It varies a lot by industry, audience and ad type, so there's no single good number. Compare your ads with each other instead. If one ad's link CTR is double another's for the same audience, it's doing a better job of grabbing interest.
Daily numbers swing a lot, especially with small budgets, so judge over a week rather than a day. If it stays high, check tracking first, then CPM, CTR and your website, in that order.
Ask for spend, results, cost per result and, for online stores, ROAS, every week, with a short explanation of what changed and what they're doing about it. At Varko we report in plain words, not dashboards full of jargon. See how we work.
