Reading CPM, reach, and fill

Open any podcast media kit and you will meet the same three numbers: CPM, reach, and fill. They look simple, and each one hides a way to be misled. Learn to read all three together and you can tell a genuinely good buy from an expensive one dressed up in flattering math. Here is the decoder.

The reason these three numbers matter so much is that they are usually all you get before you commit budget. A seller's pitch is built from them, a negotiation turns on them, and your renewal decision depends on how they compared to results. Reading them well is not a nicety — it is the core skill of buying podcast ads without getting fleeced.

CPM: the price of a thousand ears

CPM means cost per mille — the price to reach one thousand listeners. It is the headline number in almost every podcast quote, and it is calculated the same way everywhere:

• CPM = (total cost ÷ impressions) × 1,000
• Rearranged for budgeting: cost = (impressions ÷ 1,000) × CPM

Typical 2025 ranges give you a feel for the market: host-read ads run roughly $25 to $50, pre-recorded and programmatic inventory $18 to $30, and run-of-network as low as $5 to $20. By placement, pre-rolls average about $15, mid-rolls $15 to $30 and higher, and post-rolls $5 to $10.

The CPM math, worked

Concrete numbers make this stick. Suppose a show offers a $28 mid-roll CPM and averages 40,000 downloads per episode:

• 40,000 ÷ 1,000 = 40 CPM units
• 40 × $28 = $1,120 per insertion
• A four-episode flight = $4,480

Now compare it to a bigger show at a "cheaper" $18 CPM with 120,000 downloads: that is $2,160 per insertion, but if only a fraction of those listeners are your buyer, your effective cost per relevant listener may be far worse than the pricier, tighter show. CPM tells you the price of attention. It says nothing about whether it is the right attention.

CPM is the price of attention. It tells you nothing about whether it is the right attention — that is what reach and fit are for.

Reach: how many actual people

Reach is the number of unique listeners who heard your ad at least once. It is not the same as impressions: one loyal listener who hears your ad three times is three impressions but one reach. Two subtleties matter for buyers:

• In podcasting, the unique audience is often measured at the household level, not the individual, because one device may serve several people.
• Reach trades off against frequency — how many times the average person hears you. On a fixed budget, chasing more reach lowers frequency and vice versa. High reach with low frequency builds awareness; lower reach with higher frequency drives action.

When a seller quotes a big number, ask whether it is impressions or unique reach. The two can differ by a wide margin, and only one tells you how many distinct people you touched.

Fill: how much inventory actually sells

Fill rate is the share of a show's available ad slots that are actually sold and delivered. It is really a fact about the seller's business, but it affects you two ways. A high fill rate signals a healthy, in-demand show — but it can also mean a listener wades through many ads per episode, raising clutter and fatigue. A lower fill rate can mean your message has more room to land. Neither is automatically good; what you care about is whether your placement is isolated or stacked against six others.

Fill also quietly shapes price. A show running near 100% fill has little reason to discount, while a show with unsold inventory may cut you a deal or throw in added value to move slots that would otherwise expire worthless. When you sense a seller has spare inventory, that is your moment to negotiate — on CPM, on added episodes, or on a better position in the ad break.

Why all three must be read together

The temptation is to optimize for whichever number the seller emphasizes, but a good buy is a balance of the three, not a maximum of any one. Each number lies on its own:

• A low CPM looks like a deal until you learn the reach is padded with duplicate impressions.
• Big reach looks impressive until you learn most of it is the wrong audience.
• High fill looks like validation until you learn your ad is buried in an ad break of eight.

Read together, they triangulate: a fair CPM, honest unique reach among the right people, and a fill level that gives your ad space to work. That combination is a good buy. Any single number in isolation is a story the seller is telling.

Make sure you are comparing like for like

None of this comparison works unless both shows count downloads the same way. The IAB Tech Lab guidelines set the floor: at least 60 seconds of audio fetched before a listen counts, and a 24-hour deduplication window so buffering does not inflate totals. A show measuring to that standard and a show inventing its own numbers can quote the same CPM and mean wildly different things. Always ask how a seller counts before you trust what they count.

Seeing the real numbers on 10AM

The frustrating part of decoding CPM, reach, and fill is that the numbers usually arrive in a seller's PDF, framed to flatter. The 10AM Media Ads Marketplace puts IAB-aligned audience figures next to every show up front — real reach, honest pricing, and the context to compare shows side by side — and pairs them with meaning-based matching so you are weighing CPM against genuine fit, not guesswork. The three numbers stop being a negotiation tactic and start being what they should be: a clear read on what you are actually buying.