How to Lower Your CPM: 9 Proven Ways to Cut Ad Costs

To lower your CPM, broaden your targeting, refresh tired creative, cap frequency, test cheaper placements, and improve your ad relevance score. A $7.50 CPM cut to $2.50 saves $5,000 on every million impressions — and most of that saving comes from fixing targeting and creative, not from bidding tricks.

CPM feels like a price the platform sets and you accept. It is not. CPM is the outcome of an auction, and the auction rewards relevance, breadth, and freshness. Every lever below changes what the auction thinks your ad is worth — and each one is something you control. Work through them in order: the first three fix most high-CPM problems.

First, Know Your Starting Point

Before changing anything, calculate your current CPM properly: total cost divided by total impressions, multiplied by 1,000. A $750 campaign that bought 100,000 impressions has a CPM of (750 ÷ 100,000) × 1,000 = $7.50. Write that number down with the date, the platform, and the audience size — you need a baseline, or you will never know which change worked.

Then ask what the saving is worth. Cutting CPM from $7.50 to $2.50 saves $5.00 per thousand impressions — $5,000 on a million impressions. Cutting from $7.50 to $5.00 saves $2,500 per million. Those are the stakes; now the levers.

1. Broaden Your Targeting

The single fastest way to lower CPM is to widen the audience. Narrow targeting — a 50,000-person audience with five layered interests — creates a small, expensive auction. Fewer impressions exist, more advertisers want them, and the price rises.

  • Remove one or two interest layers and let the audience grow to 500,000–2 million people.
  • Test a broad audience with no interest targeting at all, letting the platform’s optimisation find buyers.
  • Keep the geography and age range honest: targeting only one city when you ship nationwide wastes money.

Expect the biggest CPM drop here. Broad audiences routinely cost a third to a half less per thousand than hyper-targeted ones. Check the result with our free cpm calculator after a few days — a $7.50 CPM falling to $5.00 on the same creative is a targeting win.

2. Refresh Tired Creative

Creative fatigue is the silent CPM killer. The same ad shown to the same people for weeks gets ignored, its relevance score falls, and the platform charges you more for the same placement. The symptoms are a rising CPM with a falling click-through rate.

  • Rotate in a new image or video every 2–3 weeks for warm audiences, every 4–6 weeks for cold ones.
  • Change the hook, not just the colours: a new first line beats a new background.
  • Test 3–5 creatives at once and kill the losers fast; one strong ad can halve your CPM.

This is also the cheapest fix on the list — a new image costs nothing next to a month of overpriced impressions.

3. Cap Frequency

Frequency is how many times the average person saw your ad. A frequency of 8 means the same people saw it eight times — and the eighth view costs the same as the first while doing almost nothing. High frequency inflates CPM because you keep buying impressions from people who have already decided.

  • Cap frequency at 2–3 per week for prospecting campaigns.
  • For retargeting, 3–5 per week is usually enough; beyond that, returns collapse.
  • If frequency is above 5 and CPM is rising, expand the audience or pause the ad set — you have saturated it.

4. Improve Your Relevance Score

Ad platforms score every ad on expected engagement. High scores get cheaper impressions; low scores pay a penalty. The score is driven by click-through rate, hide/negative feedback, and landing-page experience.

  • Write the first line for the feed, not for the boardroom: a clear benefit beats a clever slogan.
  • Match the landing page to the ad’s promise — a mismatch raises bounce rates and lowers the score.
  • Remove anything that earns “hide ad” clicks: misleading claims, clickbait, irrelevant imagery.

A relevance improvement that lifts click-through rate from 0.5% to 1% can cut your effective cost per click in half even if CPM does not move — and it usually moves CPM too.

5. Test Cheaper Placements

Not all placements cost the same. Feed placements cost more than right-column, stories, or audience-network placements — but the cheaper ones are not always worse. Test them:

  • Run automatic placements for a week and compare CPM by placement in the breakdown report.
  • Keep the placements that deliver cheap, viewable impressions; exclude the ones with near-zero clicks.
  • Do not judge a placement on CPM alone: a $2.00 CPM placement with a 0.05% click-through rate is worse than a $7.50 CPM feed placement at 2%.

Remember the viewability trap: industry data shows 30–40% of display impressions are never viewable. A cheap placement full of unviewed impressions is not cheap at all.

6. Fix Your Bidding Strategy

Manual bids set too high tell the auction you will pay anything; the auction takes you at your word. Bid caps and cost caps discipline it.

  • If you bid manually, set the cap near the CPM you actually want — not double it “to be safe”.
  • Try lowest-cost bidding for a week and record the CPM it finds on its own; that is your market price.
  • Switch to a cost cap only after you know your target: capping at $3.00 when the market is $7.50 just stops delivery.

Bidding changes show up in days, not hours — give each test at least 3–5 days and a few thousand impressions before judging.

7. Avoid Seasonal Spikes

CPMs rise when everyone advertises at once. Meta CPMs for e-commerce can spike 40–60% from September through December as holiday budgets flood the auction. A $7.50 CPM in October can become $12.00 in November with no change from you.

  • Plan prospecting pushes for January–February and late summer, when auction pressure eases.
  • If you must advertise in Q4, shift budget toward retargeting and email, where you already own the audience.
  • Never judge a new campaign by its first week in December.

8. Use Video and Native Formats

Video and native formats often clear the auction more cheaply than static display, because platforms promote the formats users engage with — and because fewer advertisers produce good video, so competition is thinner.

  • A 15-second vertical video usually earns a lower CPM than the same message as a static image.
  • Native-style ads that match the feed get more engagement, which lowers CPM through relevance.
  • Keep the hook in the first 3 seconds: an unwatched video is an expensive impression.

9. Retarget Instead of Re-Prospecting

The cheapest impressions are the ones you already earned. Retargeting people who visited your site or watched your video costs less per thousand than finding strangers — and converts better, because they already know you.

  • Build retargeting audiences from site visitors, video viewers, and email lists.
  • Spend prospecting budget only on audiences that have proven they convert.
  • Compare the two CPMs side by side: if prospecting costs $7.50 and retargeting costs $2.50, every dollar moved to retargeting buys three times the impressions.

Worked Example: What the Fixes Are Worth

Start from our running example: $750 for 100,000 impressions = $7.50 CPM.

  • Broadening targeting and refreshing creative cut it to $5.00: the same $750 now buys (750 ÷ 5) × 1,000 = 150,000 impressions — 50,000 free views.
  • Capping frequency and fixing placements cut it to $2.50: the same $750 buys (750 ÷ 2.50) × 1,000 = 300,000 impressions — three times the reach for the same money.
  • On a million impressions, the full cut from $7.50 to $2.50 saves (7.50 − 2.50) × 1,000 = $5,000.

That is the prize. And none of it required a bigger budget — just a better auction position.

What NOT to Do

  • Do not chase the cheapest CPM blindly. A $2.00 CPM on an audience that never buys is $2.00 wasted per thousand. Judge CPM alongside click-through rate and cost per acquisition — our guide to CPM vs CPC vs CPA shows how the three connect.
  • Do not change five things at once. You will never know which one worked. Change one lever, wait 3–5 days, measure, then move on.
  • Do not compare across channels. A $2.00 CPM on Google Display and a $7.50 CPM on Facebook are different products. Read what a good CPM looks like in 2026 before deciding any number is “high”.

Frequently Asked Questions

How can I lower my CPM quickly?

Broaden your targeting and refresh your creative — those two fixes produce the fastest CPM drops. Then cap frequency at 2–3 per week and test cheaper placements.

Why is my CPM so high?

The usual causes are narrow targeting, tired creative, high frequency, low relevance scores, or seasonal demand (especially Q4). Work through the nine levers above in order.

Does a lower CPM mean better ads?

Not on its own. A low CPM with a near-zero click-through rate is wasted money. Judge CPM together with click-through rate and cost per acquisition.

How much can I save by lowering my CPM?

Cutting CPM from $7.50 to $2.50 saves $5,000 per million impressions. Even a $2.50 cut saves $2,500 per million. Use our free how to calculate cpm tool to price your own saving.

Does video lower CPM?

Often, yes — video and native formats tend to clear auctions more cheaply than static display, because platforms reward engaging formats and fewer advertisers compete with good video. Test it on your own account before shifting budget.

Should I use automatic placements to lower CPM?

Test them for a week and compare CPM by placement. Automatic placements find cheap inventory, but exclude any placement with near-zero clicks — cheap impressions nobody sees are not a bargain.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top