How to measure influencer marketing ROI: KPIs, tracking and formulas

To measure influencer marketing ROI, add up everything the campaign cost, attribute sales or profit using a unique link and code for every creator, and apply one formula: ROI equals attributed gross profit minus total cost, divided by total cost. For awareness campaigns, where sales are not the point, measure cost per 1,000 views and movement in branded search instead. Set both the formula and the tracking before the first post goes live, because you cannot add tracking afterwards.

Most Indian brands do not do this yet. In a 2026 vendor survey covered by Campaign India, only 13.3% of brands tied influencer marketing to formal revenue targets, and 46.4% judged performance campaign by campaign. That leaves room to win by being rigorous. This guide gives the formulas, the KPIs for each goal, the tracking setup, a worked example in rupees, the limits of attribution and a report template. It builds on our complete guide to influencer marketing in India.

The formulas, in one table

Keep these in one place. Every number in a report should come from one of them.

Two notes on the formulas. First, some reports calculate ROI on revenue, which makes ROI equal to ROAS minus 1. That flatters the campaign, because revenue is not profit. Use gross profit. Second, whichever cost you use, use the same cost everywhere: the number in the denominator should be the full campaign cost, not only creator fees.

Count the full cost first

The most common ROI error is a denominator that is too small. Include every rupee you would not have spent without the campaign.

Whether GST on fees counts in your cost depends on whether you can claim input credit. Ask your finance team how they want it treated, and keep the same treatment across campaigns so numbers compare. This is general information, not legal or tax advice. For how creator payments and TDS work, read our guide to TDS and GST on influencer payments.

KPIs by goal

Pick the goal first, then the KPIs. A campaign trying to do everything gets measured on everything and judged on nothing.

Treat these as vanity metrics: follower gains from a collaboration, likes on their own and "earned media value" calculated by multiplying impressions by an assumed rate. They are easy to report and hard to connect to a business result. A creator who gets 4,000 likes on a post that drove no purchases is not outperforming one who got 800 likes and 60 orders.

Roughly 60% of Indian consumers say they act on influencer content, for example by buying, according to the 2025 study covered by Campaign India. That behaviour is real, but it is self-reported and it only becomes your sales number if you capture it.

Set up tracking before launch: an eight-step checklist

1. Define the conversion and the window. For example: a first purchase within 14 days of a click, or 30 days of a code use. Write it into the brief.

2. Create a unique tracked link for every creator and every post. Use UTM parameters so GA4 or your analytics tool can separate them. Example:

   https://yourbrand.in/offer?utm_source=instagram&utm_medium=influencer&utm_campaign=monsoon-haircare&utm_content=creatorhandle_reel1

3. Create a unique discount code for each creator. Codes capture sales from people who never click: they see a Reel, search the brand, and type the code at checkout.

4. Build a dedicated landing page or a deep link. Match the page to the post's promise, load it fast on mobile, and show the offer above the fold.

5. Plan the click routes. Captions on Instagram posts are not clickable, so clicks come from the link sticker in Stories, the bio link, or the call-to-action button on partnership ads. Brief creators on which one to use.

6. Install the events. Make sure your pixel, GA4 and store events fire for add-to-cart, checkout and purchase, and that UTM data reaches your order records.

7. Add a post-purchase question. "How did you hear about us?" with options including "Instagram creator", "YouTube creator", "Friend or family", "Google search" and "Ad". It takes minutes to add and it is the best cross-check you will have.

8. Record a baseline. Capture two to four weeks of normal sales, branded search and direct traffic before the first post, so lifts are visible.

Our campaign brief template includes the tracking section so creators know the link, code and label for each post.

Measuring creator ads (partnership ads)

Creator posts run as ads from the brand account are the easiest influencer spend to measure, because the ad platform reports cost per click, cost per purchase and ROAS. Meta says thousands of Indian advertisers already run Partnership Ads, as exchange4media reported in March 2025. Track them as a separate line in the report, and compare their cost per click and cost per purchase against your brand-made creatives. For the mechanics, see creator ads and whitelisting.

Measuring app installs

For an app, the tracked link goes through a mobile measurement partner (MMP): a service that attributes installs and in-app events to a source. Create one attribution link per creator, route it to the right store, and record the first meaningful event, such as registration, KYC completion or first transaction. Cost per install is the headline number and the least useful one on its own. A creator who delivers cheap installs that never register costs more per customer than one who delivers fewer, better installs.

Be aware that iOS privacy rules limit how precisely installs can be matched to a source, so iOS attribution is often aggregated and delayed. Judge iOS and Android separately, and compare creators on the same platform. Our page on influencer marketing for apps and startups covers what we track from install to first event.

Measuring awareness and brand lift

Awareness is real, and harder to pin down. Use several imperfect signals together instead of one perfect one.

You will not find a published India benchmark for the cost per 1,000 views of sponsored creator content, and we do not give one. The most defensible reference is your own paid media: one published rule of thumb says a creator is competitive when cost per engaged view is within 1.5 times your paid-media cost per engaged view, and overpriced at 3 times or more. Treat it as a low-confidence guide and use it to start a conversation, not to end one. The cost-per-view method itself is described in one published rate-card guide, with illustrative examples only.

Worked example: ROAS looks fine, ROI tells a different story

This is a hypothetical D2C haircare brand. The numbers are illustrative; swap in yours.

Costs (30-day campaign)

Results in 30 days (tracked through codes, UTM links and the ad platform)

The calculations

At 30 days, a ROAS of 1.5 sits below the break-even of 1.8, so the first-order ROI is negative. Many teams stop there and conclude that influencers do not work. The better question is what a new customer is worth. Assume, and then verify from your own cohort data, that the average new customer places 1.6 orders in six months.

The same campaign reads as −17.5% at 30 days and about +22% at six months, and nothing about it changed except the time window and the repeat assumption. That is why reports should state the window, and why repeat purchase belongs in D2C influencer plans. It is also why long creator relationships can earn more: about 62% of surveyed brand professionals agree long-term creator partnerships generate stronger ROI than short-term campaigns, per the same Campaign India coverage of the 2026 survey, which is a view, not a measured result.

Where attribution breaks

Be explicit about what your numbers miss. Four distortions matter most.

• Undercounting. People see a Reel, then search your brand, or buy on a marketplace, or ask a friend to order. None of this carries a UTM tag. Last-click tools credit it to search, direct or retargeting.
• Overcounting. Codes get shared on WhatsApp groups and coupon pages, so redemptions include people the creator never reached. Links also take credit for purchases the customer would have made anyway.
• Platforms you cannot see into. If you sell on marketplaces or quick-commerce apps, you will rarely get UTM data. Use a creator-specific bundle or code where the platform allows it, and watch branded search and rank for the product.
• Offline. For stores and restaurants, ask at the till and use a code the cashier can enter.

Three habits keep the numbers honest. Cross-check codes and links against the post-purchase survey: if the survey says 25% of orders came from creators and codes capture 10%, you are undercounting. Compare the campaign weeks against the baseline you recorded. And for large budgets, run a geo or time split: promote in one region and not in a comparable one, then compare sales.

What to put in the report

A good report fits on two pages and answers four questions: what was delivered, what it cost, what it produced, what we do next.

A leaderboard row looks like this:

The rows are made-up examples to show the layout. Our closing reports rank every creator this way, and we use the same ranking to decide what to run as creator ads. The process is described on our how we work page, and we share a live tracker from the first post.

A measurement calendar

Common mistakes

• Choosing the KPI after the results. Fix goals and targets in the brief.
• Reporting ROAS without margin. A ROAS number means nothing until it is compared with break-even.
• Using one creator's result to judge a tier. Compare tiers across several creators; see micro vs macro influencers.
• Paying on views without checking the audience. Views from the wrong geography do not sell. Our guide to spotting fake followers shows what to check.
• Planning measurement after launch. Tracking added late misses the first days, usually the best days.

If you want a tracking plan and a report format built into your next campaign, send us a brief. Our performance influencer marketing service is built around codes, links and cost per result.