Calculating customer acquisition cost across paid and owned channels
A film studio spends $4 million on paid social to promote a theatrical release and books 800,000 trackable ticket sales. The same studio sends one notification to its 12 million app users, at near zero incremental media cost, and drives 300,000 of the same bookings. Read the paid line alone and acquisition looks expensive. Blend the two and it looks cheap. The decision in front of you is not which number to believe: it is where each dollar of cost lands in the numerator, and which conversions you are entitled to put in the denominator.
What CACCACCustomer Acquisition Cost (CAC) is the total sales and marketing spend divided by the number of new customers gained in a period. It measures how efficiently you grow.View full definition → actually measures
CAC is the total cost to acquire one paying customer over a defined period and channel.
CAC = Total acquisition spend / Number of new customers acquiredThe whole argument sits in the two definitions. "Total spend" can mean media only, or media plus creative production, plus agency and platform fees, plus a share of marketing salaries. "New customers" depends on which conversion event you count, and the funnelfunnelThe customer journey from awareness to purchase, typically Awareness, Interest, Consideration, Decision, Action, with prospects narrowing at each stage.View full definition → lesson has already settled what counts as a conversion at each stage. Pick one basis and hold it for four quarters. A CAC that moves because someone quietly changed the definition is worse than no CAC.
For US streaming, planning ranges of roughly $50 to $150 per new paid subscriber circulate widely (estimate, varies by service, genre and year). Mobile games quote blended CAC anywhere from $2 to $30 per install, hypercasual at the bottom, mid-core RPGs (role-playing games) at the top (estimate, trade press). Both ranges are unusable until you know the cost basis behind them.
Paid channel CAC: the easy case
- Ad spend: $4,000,000
- Attributed bookings (via UTM-tagged links, i.e. Urchin Tracking Module parameters that let analytics trace a click back to a campaign): 800,000
- CAC = $4,000,000 / 800,000 = $5.00 per booking
That figure is auditable, which is why it dominates dashboards. Two adjustments most teams skip. First, creative: add $600,000 of shoots and cutdowns and the true paid CAC is $5.75, a 15 percent move that no media optimisation will ever recover. Second, deduplication. Meta and Google both sell the inventory they report on, and both will claim a conversion under their own attributionattributionA framework for assigning credit to the touchpoints that contributed to a conversion, so you can measure which channels and interactions actually drive results.View full definition → windows. If a Meta view-through and a Google search click each bank the same booking, your attributed totals exceed your actual transactions and every channel CAC in the deck is understated. Reconcile the sum of attributed conversions against the transaction log before anyone presents a number.
Owned channel CAC: the deceptive case
Owned channelsOwned channelsMedia channels a company owns and controls directly, such as its website, blog, newsletter, social accounts and mobile app. No per-use payment to a publisher is required.View full definition → (app push, email, the studio's own site) carry no media buy, which is not the same as carrying no cost.
- Direct media cost: $0
- Fully loaded cost: $40,000 (CRMCRMCustomer Relationship Management: software and strategy to manage and analyse customer interactions throughout their lifecycle.View full definition → platform fees, i.e. Customer Relationship Management software, plus marketing ops time to build and QA the send)
- Bookings driven: 300,000
- CAC = $40,000 / 300,000 = $0.13 per booking
Against $5.00 this looks like free money. It is not comparable. Those 12 million app users were acquired at some historical cost, usually paid, which the line ignores entirely. And a push works on people who have already converted once, which a cold impressionimpressionThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.View full definition → does not.
The harder issue is incrementalityincrementalityThe share of results (sales, conversions, revenue) that only happened because of a marketing action, not what would have occurred anyway.View full definition →. Some share of the 300,000 would have booked anyway. Hold out 5 to 10 percent of the push audience, send them nothing, and you have the answer inside a week. If half the bookings were going to happen regardless, real owned CAC is $0.26 and the channel's contribution is half of what the dashboard claims.
Platform fees belong in the calculation
For subscription apps, the store commission is an acquisition cost wearing a revenue costume. Apple and Google both charge a headline 30 percent, falling to 15 percent for small developers and for subscription revenue after twelve continuous months. A $9.99 monthly subscriber sold through in-app purchase returns roughly $7 in year one. Rule changes in the EU under the Digital Markets Act and in the US following the Epic v. Apple litigation have opened external payment links for some apps, but the default for most media apps is still store billing.
Pick one treatment and document it: either subtract commission on the value side, or add first-year commission to CAC as an acquisition toll. Doing both double counts, and teams do it more often than they admit, usually because finance nets the fee out of revenue while marketing loads it into the numerator.
The failure mode is mixing billing routes. A $70 CAC on web signups and a $70 CAC on iOS signups are not the same deal: the iOS cohort surrenders 30 percent of first-year revenue on top. Report CAC by billing route, or your cheapest-looking channel quietly funds a third party. This arithmetic, not marketing preference, is why several large streaming and audio apps stripped signup out of their iOS builds and pushed new users to the web.
Why blended CAC misleads
Blended CAC = ($4,000,000 + $40,000) / (800,000 + 300,000)
= $4,040,000 / 1,100,000
= $3.67 per bookingThe average is arithmetically correct and directionally dangerous. It hides $5.00 paid and $0.13 owned, and an executive reading only the blend may cut paid budget on the grounds that acquisition is cheap. The second-order effect lands about eighteen months later: an owned audience is a stock, not a flow. It decays through churn, app deletions and consent withdrawal, and the only thing that refills it is paid or organic acquisition. Cut the top of the funnel and the blend gets worse precisely when you need it to look good.
One more denominator trap: reactivations. Winning back a lapsed subscriber usually costs a fraction of acquiring a new one, so folding winbacks into "new customers" deflates CAC without adding a single new relationship. Count them on their own line.
Report paid and owned separately, use the blend only as a trend line, and read whatever number you get against the payback horizon the lifetime valuelifetime valueLifetime Value: the total revenue (or profit) a customer generates throughout their entire relationship with your business.View full definition → lesson sets out, since no CAC is affordable or unaffordable on its own.
Knowledge check
1. A studio's paid social campaign alone shows a high CAC, but when combined with a near-zero-cost owned-channel push, the blended CAC looks very cheap. What is the main risk of relying on the blended number alone?
2. Why can two companies both report 'CAC' figures that are not actually comparable to each other?
3. A mobile game studio wants to evaluate whether its paid user-acquisition campaign is truly efficient, separate from the effect of organic word-of-mouth installs. What approach best fits this goal?
4. Select ALL correct answers about factors that can legitimately change what a reported CAC figure represents.
Select all the correct answers.
5. Select ALL correct answers about why owned-channel acquisition (e.g., a notification sent to an existing app user base) tends to show a much lower CAC than paid channels.
Select all the correct answers.
Where the number goes soft
Since Apple's App Tracking Transparency rollout in 2021, a large share of iOS conversions arrive through SKAdNetwork postbacks that are aggregated, delayed and capped in detail. Teams fill the gap with modelled conversions, which means part of your denominator is an estimate rather than a count. Say so on the slide.
In the EU, consent rates under GDPR (General Data Protection Regulation) shrink the tracked share further, while CPMs (cost per thousand impressions) in markets such as Germany, France and Poland tend to sit below US levels. European CAC therefore reads lower for two unrelated reasons, one real and one measurement artefact. Attribution methodology alone can swing a reported CAC by 20 to 30 percent (estimate), so a pan-European figure set against a US one is directional at best.
A quick reference table
| Channel type | Typical cost basis | Attribution clarity | Scalability |
|---|---|---|---|
| Paid social/search | Media spend + creative | High (platform-reported, by the seller) | High, but rising CPMs cap efficiency |
| Owned (app push, email, CRM) | Ops/tooling cost only | Very high (first-party data) | Low, capped by existing audience size |
| In-app purchase route | Store commission, 15 to 30 percent | High | High, but margin is taxed |
| Organic/earned (press, word of mouth) | Indirect (PR team cost) | Low, hard to isolate | Unpredictable |
🎬 [VIDEO: "Customer Acquisition Cost Explained" - youtube.com/results?search_query=customer+acquisition+cost+explained+marketing - search for current marketing-metrics explainer channels breaking down CAC formulas with worked examples]
Key Takeaways
- Specify the cost basis (media only vs fully loaded) and the conversion counted before quoting any CAC, yours or a benchmark's. Creative production alone moved the worked example 15 percent.
- Deduplicate attributed conversions against the transaction log. When platforms that sell the inventory each claim the same booking, every channel CAC comes out too low.
- Owned CAC is not free and rarely fully incremental. A 5 to 10 percent holdout tells you how much of it you were buying anyway.
- Put app store commission somewhere explicit, once. A $9.99 iOS subscriber returns about $7 in year one, so CAC by billing route beats a single company-wide figure.
- Never headline the blend. It masks the paid line, it flatters you while your owned audience decays, and it collapses further if winbacks are counted as new customers.