Calculate your target CPA (Google/Meta/LinkedIn): the roadmap to copy + instructions

You can have a great CTR, “rattling” ads, a clean landing page... and still lose money. The reason is almost always the same: CPA is “driven by feeling”, not based on business reality.

The objective here is simple: set a target CPA (and, by extension, a maximum CPC/CPM) that holds up, and that adapts to Google, Meta and LinkedIn without redoing a model each time.

Summary 5 points

• Your target CPA does not leave the platform: it is deduced from what a conversion is really worth (sale, qualified lead, demo...) once the margin and variable costs are taken into account.
• An “acceptable” CPA must be written in black and white before judging a campaign. Otherwise, everything looks like “it works/it doesn't work” depending on the mood of the day.
• From the target CPA, you automatically get a max CPC (if you know your click → conversion conversion rate), then a max CPM (if you have a CTR).
• The same reasoning works on Google, Meta and LinkedIn, provided you accept a simple truth: each platform has its logic (intention vs volume vs quality) and your hypotheses must integrate it.
• The sheet below is not an oracle: it is a guardrail. Above all, it avoids decisions “based on feelings” when the amounts start to weigh.

Copy the CPA spreadsheet

‍

How do I make a copy in Google Sheet?

1. Upload the file to Google Drive (drag and drop is all you need).
2. Open it with Google Sheets.
3. Click on File → Create a copy.

You get your version, your numbers, your tabs, and most importantly: no one “edits” the reference version anymore.

How do I use the 3-step CPA spreadsheet?

Step 1 — Assume business reality (before talking about ads)

This is the part that most teams skip, and then they spend three weeks arguing about the “right CPA.” The sheet therefore starts with simple entries, deliberately down-to-earth.

To be filled in:

• Average value of a sale (or a deal).
  • E-commerce: average basket.
  • B2B: average revenue per deal (even if it's a rough average at the start, it's already better than nothing).
• Gross margin (%) : what is left before acquisition, in a “clean” version.
• Variable costs : everything that increases when you sell more (delivery, packaging, commissions, onboarding, support, tool costs per customer, etc.).
• Profit objective : the part you want to keep, even after acquisition. It's your safe zone.
• “Lead to sale” conversion rate (only if your ad conversion is not a sale).
• Typical example: on LinkedIn you optimize a “demo request” form. It's not a sale. So you need a realistic “demo → customer” rate, even an approximate one.

What the sheet makes of it:

• It turns your “pub” conversion into Expected value.
• In B2B, a lead does not have the value of a customer: it has the value of a customer weighted by your conversion rate.
• It calculates a contribution (value × margin - variable costs).
• Then she releases a Target CPA : the acquisition limit compatible with your desired profit.

At this point, you've already won something: a red line. Not an impression. A limit.

Step 2 — Define the target CPA by platform (Google/Meta/LinkedIn)

Only then, we talk about media. The same rule applies everywhere: if you know the maximum acceptable CPA, you can go up the chain.

What you provide, by platform:

• Conversion rate click → conversion (CVR).
• The “conversion” here is the event you follow on the platform: purchase, lead, appointment, etc.
• CTR (click rate), especially useful if you want to convert your max CPC into max CPM (handy on Meta and LinkedIn).

What the sheet calculates:

• Max CPC = target CPA × CVR (click → conversion)
• Intuitively: if you convert soon after the click, you should pay little for the click. If you convert better, you can pay more.
• Max CPM = function of max CPC and CTR
• Useful when the discussion turns to “LinkedIn is expensive.” Yes, often. But “expensive” compared to what? CPM is a price. What matters is the CPM that is compatible with your CTR and your conversion.

How to read it?

• On google, the game is often closer to the intention: when it is well framed, the post-click CVR can be correct, and the CPC max becomes a real control lever.
• On Meta, you pay for an inventory, a context, a creation: the CPM gives you a quick signal, but it is the quality of the promise + the landing that makes the CPA switch.
• On LinkedIn, the temptation is to accept high costs “because it's B2B”. Sometimes it's justified, sometimes it's not. The model forces you to answer a simple question: Does the expected value allow it, yes or no?

The important point: the sheet is not trying to sell you a platform. It gives you a framework to stop comparing apples and pears.

Step 3 — Use the sheet to decide (and avoid false debates)

The leaf is most useful when there is friction: when a channel is “consuming” and the results are questionable.

Here's how it's used in real life, without folklore:

1) Your observed CPA is above the target CPA

• Two specific options:
  • improve conversion (offer, landing, proof, friction, form, speed, qualification)
  • reduce the cost of access (targeting, auctions, creations, exclusions, inventory, structuring)
• An option to avoid: “let it go and hope.” When the red line is crossed, hope is not a strategy.

2) Your CPC is too high compared to the max CPC

• The useful reflex is not to “lower the stakes” blindly. You have to look at what is Fastest to move :
  • increase the CVR (often more profitable)
  • or restructure the campaign to pay less for clicks (sometimes necessary)

3) Your CPA is “good” but the quality is poor

• Classic, especially in lead gen.In this case, the problem is not the calculation, it's the definition of conversion:
  • you optimize too high up in the funnel,
  • or you have not integrated a realistic lead → sale rate,
  • or let yourself be too broad in (too permissive form, too vague promise, lack of useful friction).

Concretely: if your “conversion” does not represent value, your CPA does not represent profitability. The sheet helps you put the conversion back in its place: an economic event, not a simple green pixel.

The limitations and assumptions of the CPA spreadsheet

This sheet is deliberately simple. It works very well for framing and deciding, but it is based on assumptions. To ignore them is to tell yourself stories with clean numbers.

• Imperfect attribution : google, Meta and LinkedIn do not measure the same (windows, views vs clicks, modeling, etc.). The “platform” CPA and the “real” CPA may differ. The sheet is used to set a course, not to settle an attribution debate.
• Conversion time : in B2B, what you pay today sometimes turns into revenue much later. If your cycle is long, the good reflex is to adjust the lead → sales rate with CRM data, even approximate ones.
• Non-uniform quality : a lead does not always have the same value. Two campaigns with the same CPA can produce completely different pipelines. If you observe this discrepancy, it's not a detail: it's a signal that the optimization event or qualification needs to change.
• LTV and repurchase : if you have recurring items (subscription, repeat purchase), reasoning only at the “first purchase” may underestimate the value. In this case, the value per conversion should include an LTV (even conservative).
• Forgotten variable costs : it's the number one trap. We put in the gross margin, we forget what happens after the sale (ops, support, returns, onboarding). The target CPA then comes out too high, and the ad becomes the scapegoat.
• Incrementality : some of the conversions would have taken place without advertising (brand, SEO, direct, retail, etc.). The sheet does not measure incremental. Above all, it protects you against the other mistake: paying more than the conversion can bear.
• “A single target CPA” is not always realistic : depending on the segments, products, areas, areas, basket, qualification, you can have several target CPAs. The sheet provides a base, then you can decline by segment if necessary.

‍