In short
- Commission-only works for a minority of creators and products. A fixed floor plus a bonus per order or install is the structure most good creators accept.
- Codes and links undercount, because many viewers watch a Reel and buy later through search or a marketplace. Add a holdout or survey to measure what tracking cannot see.
- App campaigns need one mobile measurement partner link per creator and per post, and payment on a qualified event, not on a raw install.
- Judge creators on cost per qualified customer and margin, not on code redemptions alone. A steep discount code attracts buyers who may never return.
Performance influencer marketing is a way of paying creators where some or all of the fee depends on a result: orders on a discount code, sales through a tracked link, app installs, sign-ups or leads. In India it usually takes the form of a hybrid, a fixed fee that covers the creator's time plus a bonus per outcome, because few creators with real audiences will work on commission alone.
It is the version of influencer marketing finance teams like and the one most often misread. This page covers the payment models, how to track each, what tracking cannot see and how we run it. If you are new to the channel, start with our guide to influencer marketing in India.
Five ways to pay a creator for results
| Model | Creator is paid | Who carries the risk | Works best when | The catch |
|---|---|---|---|---|
| Fixed fee | A set amount per deliverable | Brand | Awareness, launches, content for ads | Pay is unrelated to result |
| Fixed plus bonus (hybrid) | A floor, plus a bonus per order, install or tier | Shared | Most conversion campaigns | Needs clean tracking from day one |
| Commission only | A percentage of net sales via code or link | Creator | Nano creators, affiliate-style content, products creators already use | Fewer posts, weaker creative, creators drop out |
| Per action (cost per install, lead or registration) | A fixed rupee amount per verified event | Creator, within limits | Apps, lead generation, sign-ups | Fraud and low-quality installs without a quality gate |
| Milestone bonus | Extra payment when a threshold is crossed, such as the first 100 orders | Shared | Creators you want to keep motivated across months | Thresholds set too high are ignored |
One reason performance data is rare in India: a 2026 creator-platform survey reported by Campaign India found that only 13.3 per cent of brands tie influencer work to formal revenue targets and 46.4 per cent judge performance campaign by campaign. Brands that set up tracking properly learn faster than most of their competitors.
Discount codes that can be trusted
A code is the simplest tracking device and the leakiest. These rules keep it useful:
- One code per creator per campaign. Shared codes make creator-level cost per order impossible to calculate. Short, speakable codes work better in video than long strings.
- First order only, one use per customer. Without this a code becomes a standing discount for returning buyers.
- A hard expiry. Match it to the campaign window and the usage rights you have bought.
- A discount your margin can survive. Work back from contribution margin (below), not from what feels generous.
- A link that applies the code automatically. Make the creator's UTM link add the code at checkout, so people who click do not lose the discount by forgetting to type it. This also lets you separate "link plus code" buyers from "code only" buyers.
- A leak check. Codes spread to coupon sites and WhatsApp deal groups. If a creator's code shows redemptions with no matching clicks from their link, you are looking at leakage, and we cap or replace the code.
A code records only the buyers who enter it. Anyone who watched the Reel and bought without it is invisible, which is why we pair codes with the lift methods further down.
UTM links and landing pages
Instagram captions do not hold clickable links, so traffic from Reels mostly arrives through the link in bio, a link sticker on Stories, or the button on a partnership ad. YouTube descriptions and pinned comments do carry links. Whatever the route, every link gets the same UTM structure so your analytics tool can group the traffic.
| Parameter | Example value | Rule |
|---|---|---|
| utm_source | The platform, spelled one way throughout | |
| utm_medium | creator | Keeps all creators in one channel group, apart from paid social |
| utm_campaign | winter-launch | One value per campaign |
| utm_content | creator042-reel1 | Creator ID plus deliverable, so you can rank creators and formats |
A full link looks like https://brand.example/product?utm_source=instagram&utm_medium=creator&utm_campaign=winter-launch&utm_content=creator042-reel1. Send creators a shortened version on your own domain, since long links look untrustworthy in a bio. Make sure your analytics tool puts "creator" traffic in its own channel instead of lumping it into direct or referral, and test every link on a phone before go-live.
Affiliate and commission arrangements
Affiliate deals run through a network or through your own tracked links. The terms matter more than the tool:
- Pay on net order value. Exclude returns, cancellations, shipping and tax. Cash-on-delivery orders in India are often refused at the door, so release commission only after the return window closes. Tell creators the payout cycle in the contract.
- Set the rate from margin, not from the market. Take the order value, subtract the discount, product cost, shipping and payment fees, and an allowance for returns. What is left, minus the margin you want to keep, is the most you can spend to win that order.
- Use an attribution window. Pick a window from click (seven or fourteen days are typical choices), write it down and apply it to every creator.
- Label it. ASCI's permitted disclosure labels include Affiliate, Collaboration and Partnership. Placement must be upfront and prominent, not buried at the end or in a bio. Our ASCI influencer guidelines explainer has the placement rules.
Hybrid fixed-plus-performance fees
A hybrid works because it splits two different risks. The floor pays for what the creator controls: ideas, shooting, editing and posting on time. The bonus pays for what depends on your product, price and checkout. We usually start from the creator's normal price for the format, keep part of it as the floor and put the rest at risk, with a cap so the maximum cost is known in advance.
| Net orders on the creator's code | Floor | Bonus at ₹120 per order, capped at ₹24,000 | Total paid | Cost per order |
|---|---|---|---|---|
| 0 | ₹18,000 | ₹0 | ₹18,000 | not applicable |
| 50 | ₹18,000 | ₹6,000 | ₹24,000 | ₹480 |
| 100 | ₹18,000 | ₹12,000 | ₹30,000 | ₹300 |
| 200 | ₹18,000 | ₹24,000 | ₹42,000 | ₹210 |
| 300 | ₹18,000 | ₹24,000 (cap) | ₹42,000 | ₹140 |
This is an illustration for a micro creator whose normal Reel price is ₹30,000, inside the typical micro range. At 100 orders the creator earns what a fixed deal would have paid; above that the brand's cost per order falls. The creator has a reason to push the code in Stories and comments for weeks, not just on posting day.
App installs and mobile measurement partners
For apps, the unit of work is the tracked link. A mobile measurement partner (MMP) such as AppsFlyer, Adjust, Branch or Singular issues one link per creator and post. It matches the click to an install and to later events in the app, then reports them to you.
- Pay on a qualified event, not a raw install. Define it in the contract: registration completed, first lesson finished, first order, trial started. Raw installs reward low-intent downloads and are the easiest to inflate.
- Hold payment for validation. A short hold, such as 7 to 14 days, lets you filter fraudulent or duplicate installs before the creator is paid.
- Use deep links. Send people to the right screen after install, not to the home screen, or the creator's best audience drops out at the first step.
- Expect an iPhone gap. On Android a tracked click can be matched to an install fairly reliably. On iPhone, Apple's privacy rules limit what any MMP can see, so attributed iOS installs are a floor.
- Test before launch. We install the app from each link on a real phone and confirm the events arrive in the dashboard.
What tracking cannot see
| Blind spot | What happens | What we do about it |
|---|---|---|
| Search later | A viewer watches a Reel, then types your brand name into Google days later | Track branded search and direct traffic around posting windows |
| Marketplaces and quick commerce | The sale happens on a platform where your code or UTM does not travel | Use the platform's tracked links where offered; rely on survey and lift |
| Offline and kirana purchases | The buyer picks it up in a shop | Survey and geo tests |
| Other devices | Watched on the phone, bought on a laptop | Accept some loss; do not over-optimise to single-device data |
| Code leakage | A code posted on a deal group is used by people who never saw the creator | Leak checks, first-order limits and caps |
| Last-click bias | A retargeting ad takes credit for a sale a creator started | Compare creator-assisted paths, not only last click |
The practical rule: sales on codes, links and installs are a floor, never the whole story.
Incrementality: proving the sales would not have happened anyway
Four designs fit influencer budgets:
- City or pin-code split. Run creators in selected cities and hold others back, then compare sales growth against matched cities. It works when you have enough daily orders per city; with low volume it is noise.
- Staggered rollout. Release creators in waves and watch baseline sales and branded search around each wave against the weeks before. Cheaper and weaker, but better than nothing.
- Holdout on creator ads. When you run creator posts as partnership ads, keep a holdout audience or run the platform's conversion-lift study.
- Post-purchase survey. One question, "How did you first hear about us?", with creator names as options, often shows buyers your tracking missed.
The KPI table
| KPI | Formula | What it tells you | Trap |
|---|---|---|---|
| Cost per click | Creator cost ÷ link clicks | Quality of traffic from each creator | Cheap clicks that never convert |
| Landing-page conversion | Orders ÷ sessions from the creator's UTM | Whether the creator's audience matches your buyer | A slow page hides a good creator |
| Cost per new customer | Floor + bonus + rights ÷ first-time buyers | The number to compare with your allowable acquisition cost | Counting returning customers as wins |
| ROAS | Attributed revenue ÷ total spend (creators plus media) | Revenue efficiency | Ignores margin and returns |
| Cost per qualified install | Spend ÷ installs reaching the key event | App campaign efficiency | Raw installs flatter weak creators |
| Code share of orders | Orders on codes ÷ all orders in the window | How much of the lift tracking captures | Rising share can mean leakage |
| 30-day repeat rate | Customers reordering ÷ customers acquired | Whether codes bought loyal customers | Low rates flag discount hunters |
Costs
A performance structure does not make a creator cheaper; it moves risk between you and them. Creator fees follow the published asking ranges: about ₹8,000 to ₹80,000 for a micro Reel, ₹50,000 to ₹3.5 lakh for a mid-tier Reel, and ₹15,000 to ₹60,000 for a micro YouTube integration, with usage rights for creator ads adding roughly 30% to 50%. Agency fees, in published benchmarks, run about 15% to 25% of creator spend. Tracking tools such as an MMP or affiliate platform carry their own licence fees, separate from creator fees. See what creators charge in India and test scenarios in the influencer cost calculator; our pricing page shows how we bill.
How we run it
- Brief. Beyond the usual audience and budget, we ask for average order value, margin, return rate, allowable cost per customer and your analytics and MMP set-up. These decide the fee model.
- Creator plan within 48 hours. The shortlist marks which creators suit hybrid deals, which suit fixed fees and which are tests, with a proposed structure for each.
- Vetting with the Gully Score. Buyer fit (20 points) and engagement quality (25 points) carry extra weight here, because a creator with great reach and the wrong buyers produces clicks and no orders. See how we work.
- Contracts and ASCI disclosure. The contract defines the qualified event, the attribution window, the payout date after the return window, the label (Affiliate, Collaboration or Partnership) and TDS treatment. Read influencer contracts in India.
- Approvals and test purchases. We place a test order on each code and link and a test install from each MMP link before anything goes live.
- Live and creator ads. The best converters become partnership ads, often the cheapest source of results in the campaign. See also UGC content creators for ad-ready assets.
- Weekly reporting. Creator-level cost per qualified result, code share, repeat signals and holdout results, with a plan to rebook, renegotiate or drop.
When performance-only is the wrong call
Avoid it for a brand nobody has heard of, where the first job is trust and recognition; for high-ticket or long-cycle purchases; for products with thin margins or unreliable stock; and for category launches where a first purchase needs several exposures. In those cases we run a fixed-fee base of micro-influencer content and add performance incentives only on the final step. For app and startup specifics, see influencer marketing for apps and startups, and for store-based brands influencer marketing for D2C brands.
To see a creator plan with a fee model for each name, send us a brief.
Frequently asked questions
What is performance-based influencer marketing?
It is an arrangement where some or all of a creator's fee depends on a measured result: orders on a discount code, sales through a tracked link, app installs, sign-ups or leads. Most Indian deals are hybrids, with a fixed fee for the creator's time and a bonus for outcomes. Our influencer marketing guide covers where this fits among other models.
Will good influencers work for commission only?
Rarely. Creators with real audiences carry the cost of the shoot and edit, and they take on risks they cannot control: your price, your stock and your checkout. Commission-only suits creators who already sell your category, nano creators who value the product, or affiliate-style content. For most campaigns a floor plus a bonus per outcome gets better creators and steadier posting.
How do you track sales from influencer discount codes?
Give each creator a unique code per campaign, limit it to first orders, and read redemptions from your store or app backend. Pair it with a UTM link that applies the code automatically. Codes only capture buyers who enter them, so treat the count as a floor and compare it with a survey or lift test.
How do you track app installs from influencers in India?
Through a mobile measurement partner such as AppsFlyer, Adjust, Branch or Singular. You create one tracked link per creator and post, and the partner matches clicks to installs and in-app events like registration or first purchase. On iPhone, Apple's privacy rules limit attribution, so read iOS numbers as a minimum.
Is influencer marketing the same as affiliate marketing?
No. Affiliate marketing is a payment model: the partner earns a commission on tracked sales. Influencer marketing is a channel that can use that model, a fixed fee, or both. An affiliate link in a creator's post is still an ad and needs an ASCI-compliant label such as Affiliate or Collaboration.
How do I measure ROI when tracking misses sales?
Use three layers: tracked sales (codes, links, installs), a post-purchase survey asking how buyers heard of you, and a lift test such as a city split or a staggered launch. Together they bracket the true effect. We walk through the method in how to measure influencer marketing ROI.
How much does performance influencer marketing cost?
Creator fees follow the same published ranges as any campaign, for example about ₹8,000 to ₹80,000 per micro Reel. A performance deal changes who carries the risk, not the underlying price of a creator's work. See influencer rates in India and plug options into the cost calculator.