Influencer cost per acquisition: CPA maths, break-even and worked examples

Influencer cost per acquisition is the total cost of a campaign divided by the acquisitions it produced: orders, new customers, qualified installs, leads or enrolments. The number only means something beside break-even CPA, the most you can pay for one acquisition and still make money. Work that out first from contribution margin, count every cost line, and compare both on a first-order and a six-month basis.

This guide gives the formulas, a worked break-even and campaign example, a sensitivity table, the sample-size reason to be careful with creator-level CPA and short examples for apps, lead-led brands and restaurants. All the figures are placeholders labelled as examples, because we found no published India benchmark for CPA or conversion rates on creator campaigns. It builds on the CAC row in our guide to measuring influencer marketing ROI. Few teams start from an allowable CPA: only 13.3% of surveyed Indian brands tie influencer work to formal revenue targets, per a 2026 vendor survey covered by Campaign India. For the wider plan, see the complete guide to influencer marketing in India.

Pick the acquisition first

CPA changes with what you count. Write the definition into the brief.

What goes into the cost

A CPA built on creator fees alone flatters the campaign. Use a fully loaded cost.

The discount needs one rule: count it once. If your contribution per order already starts from net revenue after the discount, do not also add the discount to cost. If you start from gross revenue, add it as a cost line. Doing both makes every campaign look worse than it was.

Break-even CPA: the number to compare with

This example uses placeholder numbers for a hypothetical D2C brand. Swap in yours.

That ₹600 is the first-order break-even CPA. Spend more than that to win a customer and you lose money on the first order. If you want to keep 30% of the contribution as profit, your target is ₹420.

For a longer view, add repeat purchases. Assume, then verify from your own cohort data, that the average new customer places 0.4 further orders in six months. Repeat orders carry no code, so each contributes ₹1,400 minus ₹660 of costs, which is ₹740. That adds 0.4 × 740 = ₹296, so six-month contribution is ₹896. The same 30% rule gives a six-month target of about ₹627. We found no India benchmark for repeat rates, so do not borrow one: use your own.

A campaign, worked through

The campaign has 12 micro Reels at ₹25,000 each. That fee sits inside the ₹8,000 to ₹80,000 micro range in our rates guide.

The campaign produced 1,200 tracked orders, of which 900 were from new customers. The other 300 were existing customers using a code.

Cost per order looks profitable. CAC says the first orders lost a little, and the campaign pays back only if the repeat assumption holds. The gap comes from the 300 existing customers: their orders count as revenue but not as acquisitions. Quote CAC as the headline. Our post on tracking sales with discount codes shows how to separate them.

What would change the answer

Tracking is never exact. Here is how the CAC moves if the errors are real. These are what-ifs, not estimates, because we found no published India figure for either error.

The last row is the point: the errors can cancel, so you cannot assume a direction. That is why we report a range and cross-check with a survey, as set out in our post on influencer marketing attribution.

Average CPA and marginal CPA

The average CPA of a campaign hides what the next rupee buys. As you add creators, the best audiences are used first and each new creator tends to bring customers at a higher cost. What matters when you scale is the marginal CPA: the cost of the extra customers from the extra spend.

Suppose a second wave adds ₹3,00,000 and 380 new customers to the first wave's 900.

The blended figure of ₹684 looks like a small slip. The second wave on its own cost ₹789 per new customer, above the first-order break-even of ₹600 and only just inside the ₹896 six-month break-even, and that holds only if repeat purchases do. The decision rule is simple: keep adding spend while the marginal CPA stays below your allowable CPA, and stop or change the mix when it does not. Without a marginal view, you scale a campaign past the point where it earns money and never see it in the average. The numbers are made up, and we found no India data on how fast CPA rises with scale, so measure yours wave by wave.

Creator-level CPA: small numbers mislead

Counts of orders fluctuate by chance. As a rough guide, a count of n orders has a 95% range of about plus or minus 1.96 ÷ √n of itself.

A creator paid ₹30,000 with 5 tracked orders shows a CPA of ₹6,000. On another day the same post could have produced 3 orders (₹10,000) or 8 (₹3,750). Ranking creators on that number is ranking noise. Our working rule, not a published threshold, is not to rank a creator below about 25 tracked orders. Until then, pool creators by tier, format or language, and label small samples "test again" in the campaign report.

A feasibility check before you book

CPA can be turned around: how many orders must a post bring to justify its fee? Divide the fee by your break-even CPA, then multiply by your loaded-cost ratio. In the example above, all-in cost was 1.92 times creator fees because it includes ad media; yours will differ.

Set that beside what the creator's audience can plausibly deliver, using your own click-through and landing-page conversion rates, as in the work-backwards method in our KPI guide. If a post needs 256 orders and a creator's link will realistically produce 40, the fee is wrong or the product is the wrong match. The influencer cost calculator helps with the fee side.

The same maths for apps, leads and restaurants

These are placeholder examples to show the method, not benchmarks.

In each, the cheap number (cost per install, per lead, per offer) is not the decision number. Move down the funnel to the event that earns money.

Mistakes to avoid

• Counting only creator fees. Add rights, fees, product, production and media.
• Counting existing customers as acquisitions. Quote CAC, not only cost per order.
• Counting the discount twice. Net revenue or cost line, not both.
• Judging on first orders alone where repeat purchase is normal for the category.
• Ranking creators on tiny counts. Pool them or retest.
• Mixing organic creator posts and creator ads in one CPA. They answer different questions.
• Quoting a CPA as an industry standard. There is no India benchmark to quote.

If you want a break-even and a CPA target worked out for your product before the first brief, send us a brief. Our performance influencer marketing service is built around that number, and our micro-influencer service is where many first-wave tests begin.