Shopify Collabs tracks affiliate sales through each creator’s unique link and discount code, and it uses those to calculate and pay commissions. That tracking is good enough to pay creators fairly. It is not the same as knowing which creators bring in customers you would not have gotten anyway. To get that, you need to read Collabs data next to your attribution tool and account for the overlap between codes, links, ads, and email.
How does Shopify Collabs track sales?
Collabs is Shopify’s own affiliate app. You recruit creators through an application page or direct invites, give each one a personalized link and code, and Collabs records sales that come through either and pays commissions through your Shopify bill. It is free to install, and you pay commission when sales happen, plus any additional charges Shopify lists. Because it runs inside Shopify, its order matching is reliable. The question is not whether it recorded the order. It is whether the creator caused it.
Why do Collabs numbers look better than they are?
Codes leak. Within weeks, a creator’s code is often on coupon sites and browser extensions, and shoppers who were going to buy anyway find it at checkout. Collabs will credit that creator for the sale, and you will pay commission on it.
Credit overlaps. A shopper might click your Meta ad, get a Klaviyo email, and then buy with a creator’s code. Meta, Klaviyo, and Collabs will each count that order. Add them up and you will have more attributed revenue than total revenue. I wrote about the email side of this in the problem with last-click attribution.
Some sales were already coming. Your existing customers follow creators too. If a repeat customer uses a code, Collabs counts it the same as a brand new customer.
How do you get a truer read?
Split new and returning customers. Look at how many of each creator’s orders came from first-time buyers. A creator whose sales are mostly repeat customers is a discount channel, not an acquisition channel.
Check where the code was used. If a creator’s code shows up on orders whose sessions came from coupon sites or direct traffic with no link click, treat those as leakage. Your attribution tool can show the session source for code orders if it tracks discount codes.
Make links do the work. Check whether the links Collabs generates for your creators carry UTM parameters in your setup. If they do not, your attribution tool will see those visits as referral or social traffic with no creator attached, and you will have to lean on codes alone.
Read Collabs next to an independent tool. ThoughtMetric tracks influencers and affiliates by UTM parameter or discount code and can report orders, revenue, and net sales by code alongside the channel that drove the visit. (Disclosure: ThoughtMetric sponsors this site.) Any attribution tool that does this lets you see a creator’s orders in the context of every other touchpoint, rather than only through the commission ledger.
What should you pay creators on?
That is a business decision, not a measurement one, but measurement should inform it. Many brands pay commission on all code and link sales because it is simple and creators expect it. That is fine. Just do not use the same number to decide which creators to scale. For scaling, use new customer orders from link clicks plus a share of code orders you trust, and check it against post-purchase survey answers that name the creator. More on influencer measurement in Shopify attribution tools with post-purchase surveys.
What about returns?
Check how your commission terms handle refunds, and read each creator on net sales, not gross. A creator whose audience returns a lot can look like a top performer on total sales and an average one after refunds. If your attribution tool reports net sales and refunds by discount code, compare that view with what Collabs shows before you raise anyone’s commission rate.
How often should you review it?
Monthly is enough for most programs. Pull each creator’s orders, new customer share, refund rate, and commission paid. Rank them on new customers per dollar of commission, not total sales. In most programs a few creators drive most of the real growth, and the rest are mostly paying out on sales you would have made anyway.
A simple checklist
- Give every creator a unique link and code, and expect the code to leak.
- Split each creator’s orders into new and returning customers.
- Check whether Collabs links carry UTM parameters, and add them where you can.
- Read creator results in an attribution tool, not only in Collabs.
- Running creator content as paid ads too? Measure those separately, as in how to measure Meta partnership ads.
- Pay on Collabs numbers if you like, but scale on new customers per dollar of commission.
- Ask buyers how they heard about you and look for creator names in the answers.
Leave a Reply