flowpeekresources / guides / Marketing
How to measure influencer sales: code, link and time window
You'll plan how to measure an influencer collaboration, with the code, link and time window set before publication, and then account for sales and the cost per new customer.
Short answer
The key answer
Combine three methods. Give each creator their own discount code and their own UTM link, and before the post goes live, set the time window in which you count the effect. The code shows purchases by people who didn't click the link, the link shows visits and what happened next, and comparing sales with a baseline period shows part of the effect that left neither a code nor a click. None of these numbers alone is the full result, so write the counting rules into the brief or contract.
- Code, link and time window set before publication
- Sales without counting the same order twice
- Customer acquisition cost and the decision about the next collaboration
Key terms in plain language
- Time window
- A period agreed in advance after publication, during which you count orders attributed to the collaboration.
- Assisted conversion
- A purchase where the source appeared on the customer's path, even though it wasn't the last touchpoint.
1. Set the counting rules before publication.
It's easy to argue about the results of a collaboration when the rules are set after the fact. Before publication, write down in the brief: the goal of the collaboration, the code and link assigned to the creator, the time window in which you count orders, and how you count orders that have both the code and a visit from the link. Also record the costs: the fee, the value of products provided and the cost of the discount.
Choose the time window based on the format and the product price. An Instagram story is visible for 24 hours unless the creator adds it to their highlights, while a video or post can attract viewers for weeks. Customers buy a product costing a few dozen zlotys faster than equipment costing several thousand. More important than the length of the window itself is that it's the same for all the creators you compare. Also remember to label the collaboration: according to UOKiK, the Polish consumer protection authority, influencers, advertising agencies and advertisers are responsible for clearly labeling advertising.
| Brief item | What to write down |
|---|---|
| Goal | Sales, new customers or reaching a new audience |
| Code | One code per creator, with an expiry date and a per-customer usage limit |
| Link | A URL with UTMs for each post |
| Time window | From the publication date to a set date, the same for the creators being compared |
| Order with a code and from the link | Counted once, according to the written rule |
| Costs | Fee, products, discount cost and handling |
2. Give each creator their own code.
A discount code captures purchases by people who saw the post but reached the store some other way than through the link: they typed the address, found the store in search or came back a few days later on another device. That's why each creator should have their own code, not a shared campaign code. You'll find orders with a code in your store's admin panel or in an order export that includes the discount used.
Codes have limitations too. A code can end up on coupon sites and be used by people who never saw the post. Some viewers will buy without the code because they forget it. Set an expiry date and a per-customer usage limit, and review orders with the code for the source of the visit. If many of them come from coupon sites, part of the result doesn't come from the creator.
3. Add a UTM link to every post.
The link shows how many people came from the post, what they viewed and whether they bought. Prepare a separate URL for each creator and format, e.g. utm_source=instagram, utm_medium=influencer, utm_campaign=autumn_2026, utm_content=creator_name_story. Stick to the same naming dictionary as in your other campaigns. If you report in Google Analytics 4 and want to see collaborations as a separate channel, create a custom channel group.
In many formats the link isn't clickable, e.g. in the caption of a regular Instagram post. In that case, the creator points viewers to the link in their bio or in a story. Some people will type the store address manually anyway, so the link won't show all the visits or purchases the post contributed to. Before publication, open the link on a phone and check that the parameters survive any redirects.
4. Count sales within the window and compare them with a baseline period.
When the window closes, collect three numbers: orders with the creator's code, orders attributed to visits from their link, and orders that have both. Count the last group only once. The total without double counting is the sales you can directly connect with the post.
Part of the effect leaves neither a code nor a click. To estimate it, compare total store sales, direct visits and brand name searches during the collaboration window with a comparable period without a post. A rise above the usual level may come from the post, but also from the weather, a promotion or another campaign. Treat it as a hint, not a result to settle up on, and don't automatically add it to the creator's sales.
| Item | Worked example |
|---|---|
| Orders with the creator's code | 38 |
| Orders from link visits | 22 |
| Orders with both the code and the link | 12 |
| Sales connected with the post | 38 + 22, minus 12 counted twice = 48 orders |
| Rise in direct visits above the usual level | A hint for the write-up, outside the settlement |
5. Check what role the post plays in the path to purchase.
A creator's audience is often made up of people who are only just discovering the brand. They may buy a week later, after clicking a search ad. In a last-touch model, the ad then gets the entire order and the collaboration doesn't show up at all. That's why, when evaluating influencers, you should also look at the first-touch model and assisted conversions, i.e. purchases where a given source appeared earlier in the path.
In Flowpeek, you can compare the same set of orders across several attribution models with a chosen window, and see assisted conversions separately from attributed sales. In the screenshot shown, the "ig" source has 1 conversion in the First Touch model and 7 in Last Touch. The same channel looks different depending on the model, so when settling a collaboration, write down the model and window you use.
6. Calculate the cost per new customer and decide on the next collaboration.
Add up the collaboration cost: the fee, the value of products given to the creator, the cost of the discount granted on orders with the code, and handling time. Divide it by the number of orders connected with the post, and separately by the number of new customers among the buyers. You can check new customers in your store data: they're people who haven't placed an order before.
Worked example: a collaboration costs a PLN 4,000 fee, PLN 400 in products and PLN 1,200 in discounts, PLN 5,600 in total. With 48 orders, the cost per order is about PLN 117, and if 28 of them were placed by new customers, the cost per new customer is PLN 200. Compare these numbers with the acquisition cost in other channels and with what you keep from an order. After two or three months, check how many new customers bought again, and only then decide on the next collaboration.
Common mistakes
One code for all creators
A shared code makes it impossible to compare collaborations. Each creator needs their own code and link.
Counting orders twice
An order with a code that came from a link visit is a single order. Count it once according to the written rule.
Judging only with a last-touch model
A post can start the customer's journey. Check first touch and assisted conversions before you write off a collaboration as ineffective.
For your next review
Checklist
- Each creator has their own code and their own UTM link.
- The time window and the way orders are counted are written down before publication.
- The collaboration is labeled as advertising in line with UOKiK guidelines.
- We count orders with a code and from the link without duplicates.
- The collaboration cost includes the fee, products and the cost of the discount.
Now your store
Settle one collaboration using a fixed card.
- 1
Before publication, fill in the brief: goal, code, link, time window and costs.
- 2
When the window closes, collect orders with the code and from the link, remove duplicates and check how many buyers were new customers.
- 3
Calculate the cost per order and per new customer, and after two or three months add repeat purchases.
From report to decision
Example collaboration settlement
A worked example using made-up numbers. It shows how to record the data, not a typical result of an influencer collaboration.
- Window
- 14 days from publication, the same for all creators in the campaign.
- Sales
- 38 orders with the code, 22 from the link, 12 with both. 48 orders in total, including 28 from new customers.
- Cost
- PLN 5,600: fee, products and discount. About PLN 117 per order and PLN 200 per new customer.
- Comparison
- Next to it, we record the cost per new customer from ads in the same period, calculated using the same attribution model.
- Decision
- We decide on a second collaboration after checking new customers' repeat purchases after three months.
The settlement is based on rules set before publication, so you can repeat it for the next creator and compare the results.
Questions and answers
Which is better: a discount code or a UTM link?
Ideally both. The code captures purchases without a click, the link shows visits and behavior in the store. Each method on its own misses part of the sales.
How long should the time window be?
Choose it based on the post format and product price, then use the same length for all the creators you compare. Write it down before publication.
Is the store responsible for how an influencer labels advertising?
According to UOKiK, influencers, advertising agencies and advertisers are responsible for clearly labeling advertising. Include labeling in the brief and the contract.
From knowledge to report
Open the right module.
Traffic sources
See where visitors come from, whether ads, search or e-mails, and which sources lead to orders, not just clicks.
Multi-channel attribution
Flowpeek combines its own visit tracking with your store's orders and attributes sales to channels using the model you choose. Compare six models and see which sources start, support and close a purchase.
Product analytics
Set interest in your product range against carts and order data. Find products worth giving more visibility to, and ones whose offer is worth improving.
Read the next guide.
All guidesMarketing / 9 min
UTM links for online stores: tagging campaigns and sales sources
Learn the UTM parameters, write down a naming dictionary and check links before they go out, so campaign orders don't end up under direct traffic.
Analytics / 10 min
Why do GA4, your store and your ads show different results?
Align dates, revenue definitions and attribution. Compare specific orders, find the source of the discrepancy and agree on a shared reporting rule.
Marketing / 11 min
How much should an online store spend on ads? Set your budget
Work out the maximum cost per order from your margin, plan the budget from a sales goal and scale it in steps while checking the cost of extra orders.
Free · 30 minutes · online
GA4, your store and your ads show different numbers? We'll check it with you.
We compare one month of sales across your tools and point out three fixes to your measurement. No strings attached. What we check
30 days on your own data
See where customers come from after a creator's post.
In a 30-day Flowpeek trial, you'll check visits from UTM links, what visitors do next in your store and each source's share of sales across several attribution models. We'll handle the setup within 48 hours.
- No fixed-term contract
- We set everything up for you in 48 h
- App servers in Germany (EU)
- Cancel anytime
30 days for PLN 0 · first payment after the trial · cancel anytime
See pricing after the trial