You’ve probably heard this: forget the million visits; a smaller audience with a higher CPM can earn just as much or even more. But is it true? Maybe it’s just peace of mind for small publishers?
Long story short, we checked. We pulled real numbers from our own stats, asked traffic specialists what they actually see, and we’re ready to settle this once and for all.
The clearest sign that your huge impressions volume is nothing but a vanity metric is that it keeps climbing while profit and CPM sit flat. You’ve probably seen it right in your stats: traffic spikes, but the profit barely moves.
Here is one example we took from one real publisher’s daily numbers in June 2026:
| Date | Impressions | Revenue | Costs | Profit |
|---|---|---|---|---|
| Jun 14 | 1,444,067 | $59.92 | $19.86 |
$40.06
|
| Jun 15 | 1,584,712 | $80.95 | $31.45 |
$49.50
↑ +$9.44
|
| Jun 19 | 1,717,117 | $38.17 | $17.04 |
$21.13
↓ −$28.37
|
| Jun 20 | 4,154,110 | $28.51 | $12.30 |
$16.21
↓ −$4.92
|
| Jun 21 | 5,528,867 | $33.21 | $14.77 |
$18.44
↑ +$2.23
|
| Jun 23 | 6,941,770 | $29.16 | $10.80 |
$18.36
↓ −$0.08
|
| Total | 21,370,643 | $269.92 | $106.22 | $163.70 |
Look at the last two rows. Impressions grew from 1.4 million (Jun 15) to 6.3 million (Jun 23) – it’s a 4.5x jump in traffic. Meanwhile, profit actually dropped from $45.00 to $16.69. So, the volume growth didn’t obviously help… like at all.
Why does it happen? In many cases, when you see such traffic growth without any positive movements in profits, it means that your growth comes from a zone with near-zero CPM.
Ioannis Giouroukakis, Head of Publishers at Monetag:
Simple rule: if you can’t explain a metric through revenue growth over the same period, it’s not a growth metric – it’s a volume metric wearing a growth metric’s clothes.
So, you don’t really need to care about impressions?
Yes and No.
We pulled some more data from our own network stats to test both theories – and both of them turned out to be true.
We compared two very different scales of publisher by loosely dividing them into ‘big’ and ‘small’. Here is what we got from three ‘small’ and three ‘big’ publishers:
| Publisher | Impressions | CPM | Profit |
|---|---|---|---|
| Small Publishers | |||
| Small #1 | 435,913 | $38.49 | $6,224 |
| Small #2 | 239,282 | $40.84 | $4,078 |
| Small #3 | 313,085 | $23.07 | $2,494 |
| Big Publishers | |||
| Big #1 | 42,738,022 | $0.011 | $276 |
| Big #2 | 114,220,554 | $0.006 | ~$100 |
| Big #3 | 115,530,268 | $0.000 | $0 |
Here, we don’t see any correlation between impressions and CPMs – with real, no-bot traffic from both groups. And such a comparison might persuade someone that traffic increase isn’t worth it at all, on top of having to watch for downtime and handle traffic spikes.
But here’s what we also have.
For the other half of the story, we checked the absolute top earners across the whole network. And the winner grabs a massive share of impressions:
| Publisher | Impressions | CPM | Profit |
|---|---|---|---|
|
Biggest earner
High volume |
10,000,000 | $1.50 | $16,000 |
|
Best small-scale earner
High CPM |
435,913 | $15.00 | $6,500 |
Even a modest CPM ($0.09) turned into six figures of profit for him – thanks to, obviously, billions of impressions. No small-scale publisher came close to that absolute number, even the one with a very decent CPM rate. In this case, volume definitely wins.
Or, in other words, why some low-volume publishers ultimately outearn high-volume ones, and why the reverse also exists? Here are some reasons explained by Ioannis Giouroukakis, Head of Publishers at Monetag.
The same CPM number can hide different realities. As Side A shows, two publishers can have almost the same-looking CPM but earn it in completely different ways. One gets their profits from lots of cheap impressions, another from far fewer but thanks to much higher-value actions their site visitors do.
Niche and audience match also matter. Publishers working in narrower, more specific niches like APK or anime tend to see stronger advertiser demand and better rates than broader, generic audiences.
Niche and audience match matter
Publishers in specific niches see stronger advertiser demand — and higher CPMs
Niche publishers
Generic publishers
Niche beats volume
Anime publishers earn 1.9× more per 1,000 impressions than Social in Tier 1 — with 2× fewer impressions. Advertisers pay a premium for intent-driven, targeted audiences.
Jul 2026 · Tier 1 only (US, UK, DE, AU, CA, FR) · CPM = revenue / impressions × 1,000
There must be a match between a publisher and an advertiser. Sometimes a specific traffic source just happens to match what advertisers want right now, so it earns more than a source that looks more premium on paper. What actually pays isn’t some fixed internal score; it’s more about whether real advertisers want that specific audience today.
The honest conclusion: what really matters is a combination of what’s actually behind these impressions and CPM numbers.
Traffic alone isn’t the answer; some other actions and metrics might shed more light on the whole picture.
To be more precise, your essentials are RPM, revenue per session, and returning-visitor rate.
Open your Monetag dashboard, and you’ll see CPM. That number tells you what 1,000 ad impressions paid you, but doesn’t tell you what a visitor to your site is worth. If you want to dig deeper, you need to count RPM:
An example: last month you earned $2,029 from 5,000,000 impressions. Monetag shows your CPM: $0.41. Now check your site analytics: say you had 2,500,000 page views that month. Your RPM is $2,029 ÷ 2,500,000 × 1000 = $0.81.
So one visit to your site is worth about 0.8 cents. It’s double your CPM because each visit produced two impressions instead of one. So, the entire formula in plain words:
RPM = impressions per visit × CPM
Previous
Current
| Metric | Previous | Current | Change |
|---|
For you, it means there are only two ways to earn more: show each visitor more ads, or get more money per ad. Everything you do – add a format, change frequency, filter geos – moves one of those two numbers.
Note: A falling RPM doesn’t always mean you are not doing okay. For example, according to Trade House Media, ‘a decrease in page RPM can come as a result of an increase in site traffic’. It’s not that the new traffic is cheaper: at Monetag, fresh users usually earn more, as their ad views haven’t been used up yet, but growth changes the mix: more page views per visit, or a new geo where advertisers pay less, drag the average down even if your total earnings went up. So, never read RPM alone, without considering page views and the overall revenue.
While RPM counts page views, RPS counts visits – a metric you can find in your Google Analytics. As one visit often covers more than one page, the RPS formula looks like this:
RPS = Total Revenue ÷ Total Number of Sessions
Your month again: $2,029, and 1,250,000 sessions. That’s about $0.0016 per visit.
According to Playwire, RPS provides a clearer picture when your traffic mix changes. Say users come from social media and only stay on one page during their visit, instead of your average three. Still, they bring you the same per impression. Your page views go up, and so does revenue, but RPM doesn’t move at all. Meanwhile, each visit is now worth a third less than it was, and RPS shows it.
Playwire notes that RPS also matters more when your revenue stops coming from one place: ‘For publishers with paywalls or mixed monetization models, RPS captures value more accurately than RPM alone.’
So, in short: RPM tells you if a page earns, RPS tells you if a visitor earns.
Another metric to check in your Google Analytics is returning visitors – and count their rate with the simple formula:
Returning visitors ÷ total visitors × 100%
Why do you need this? RPM and RPS can only tell you what you earned, but they can’t predict whether you’ll earn it again. Let’s look at the potential picture:
If you’d checked your returning visitor rate in month 1, you’d see it dropped – say from 40% to 34% – and this could be a warning sign. And you’d have done something different with it: like have kept the new format at a lower frequency, earning 15% more instead of 25%. Sounds worse, but in fact better in the long run.
Returning-visitor rate is the only metric on this list that moves down when the others move up, and gives you a hint.
Relentlessly chasing an increase in page RPM can convince publishers to force too many ads onto the user. […] users will increasingly employ ad-blocking extensions and discourage users from exploring other pages on the site, or from returning to your site at all.
So, the tip: check it whenever you change anything about your ads, and give it a few weeks. If revenue rose and the returning rate held, keep going. If revenue rose and the returning rate fell, try a different frequency and watch whether that section starts losing readers faster than the rest.
None of these numbers means much on its own, so read them as a set. Mark what each of yours did since your last report and see what’s actually happening. This simple checker also includes the combinations that can’t happen – it usually means your two data sources aren’t describing the same thing:
RPM Diagnostics
Mark what each metric did since your last report.
Two industry facts: a slow page loses visitors before the ad even appears, and an ad only earns full price if someone actually sees it. And what if the ad loaded, but was never fully opened because the site was lagging? It’s counted as served, but not viewed.
It sounds harmless enough, but if this happens too often, your placement gets a bad viewability track record. As a result, advertising networks might filter it out as low quality, and your CPMs drop. Meanwhile, a one-second delay in page load can cut ad views by nearly 11%.
Worth fixing? Industry estimates put the gain from lifting viewability from 40% to 70% at roughly a 30% RPM uplift, driven by exactly those bidders coming back. If you’re already at 65%, expect far less. The money is in fixing genuinely bad placements, not polishing decent ones.
So a quick tip is to solve the slow page loads, if this problem exists.
Visitors from different countries bring different CPM and profit, and Tier 1 pays more on average. However, it’s still just the average, and it doesn’t tell you exactly what you will earn. What matters more is whether your site fits the country.
Two examples:
So pick a few countries, write in their language, and check which search engine they use there. This can do more for your earnings than simply getting more visitors.
If all your visitors come from one place, say, Facebook, then Facebook decides your income. If you have three sources and one drops, you lose just a part. Yes, the same old ‘don’t put all eggs in one basket’.
Another reason is that visitors from different sources are worth different money. Here’s what MGID’s data shows:
So it’s neither the biggest audience nor the highest CPM that wins on its own. Volume and CPM only pay off together, and both of them come from the same source: how well your audience, geo, and ad formats fit what advertisers are buying right now.
Being a successful publisher means treating a site like a business: knowing your RPM by zone, knowing which audiences come back and why, and knowing exactly that ratio before you decide whether to chase more traffic or better traffic.
Yes, a million visits looks good on a screenshot – but you need more numbers to boost your paychecks.