Most affiliate creators run two or three discount codes at once: one from a long-running brand partner, one from a recent launch, one evergreen Amazon-side code. The brand dashboards tell you how many people used each code, but they don't tell you which post, which platform, or which bio placement actually sent the click. The result is the same kind of attribution fog every marketer eventually hits: you know what worked, but not why. This is how to get unstuck.
Why this matters
Two reasons creators end up wanting the answer:
- To double down on what works.If your TikTok bio converts the SAVE15 code at four times the rate of your Instagram stories, the next campaign should weight TikTok more heavily. Without per-source attribution, you can't see that signal.
- To negotiate better deals.When a brand asks how your code did, "we drove 312 redemptions, 70% from Instagram stories" is a meaningfully stronger answer than "we drove some." The platform breakdown becomes part of the renewal pitch.
The four signal sources
There are four places attribution data for an affiliate code can live. Each one tells you a different slice of the picture; the useful view is cross-referencing two or three of them.
1. The brand's redemption dashboard
Every brand running discount codes sees redemptions in their e-commerce backend: Shopify, WooCommerce, BigCommerce, or whatever platform processed the order. If your code is unique to you (SAVENAME15 with your handle in it), the brand can pull a clean count of how many times yours specifically was used.
Strengths: this is ground truth for revenue. Every redemption actually drove a sale, and the dollar value is real.
Limits: brands rarely share this in real-time. Most send a monthly or quarterly summary, and many won't share the per-post breakdown even if they have it. By the time you see "your code drove $X," the campaign is over and the attribution decisions you might have made along the way are moot.
2. Per-link click counts on the bio link
If your bio link runs through a tool that tracks clicks per destination (rather than just total page views), you get a leading indicator: which products are being tapped on, from which platform, in real time. The click count isn't the redemption, but it's a strong proxy: high-click products with codes attached usually correlate with high-redemption codes.
Pair the click data with the brand's eventual redemption report and you have the conversion-rate denominator: 1,000 clicks, 50 redemptions, 5% conversion. The next time you promote a similar product, you can predict revenue from clicks.
3. UTM-tagged URLs (where applicable)
UTM parameters (the five utm_* query strings every analytics tool reads) let you tag the exact source/medium/campaign a click came from. Hand-tag the link you put in a TikTok bio with utm_source=tiktok&utm_medium=bio; tag the Instagram version differently. When clicks attribute, you can cleanly slice by which platform fired them.
The catch: UTMs only work if something downstream reads them. Affiliate brand dashboards typically don't expose UTM-level breakdowns to you. The fix: send the click through your own redirect first (a bio platform that tracks clicks server-side before forwarding) so the UTM attribution happens on your side before the handoff. Some bio platforms attach UTMs automatically based on which platform the referrer came from; check your platform's docs.
For non-affiliate links where you control the destination (a brand's direct landing page), UTMs work end-to-end and the brand's own analytics can confirm the source.
4. Customer-side feedback
The lowest-tech signal source, still useful: ask your followers where they heard about the code. A line in a story ("reply with the post that sent you here"), a Q&A widget, a one-question survey on the link-in-bio page. The data is noisy because not everyone responds, but the responses tend to be directionally honest.
This signal source is most valuable when the other three give you ambiguous answers. If your dashboard shows 200 redemptions but you don't know which post sourced them, ten qualitative replies on a story can break the tie.
Cross-referencing the signals
Each signal alone has gaps; together they cover the question. The pattern that actually works:
- Click counts (real-time) tell you which surface is doing the work day by day. Refresh weekly; reweight your content if a clear winner is emerging.
- Redemptions (monthly)from the brand confirm the conversion rate per surface. Multiply the conversion rate by the upcoming month's expected clicks to project revenue.
- UTM-tagged data (when available)resolves ambiguity when the click count alone can't (e.g. a stories drop and a feed post both ran the same week; UTMs tell you which one moved the needle).
- Audience feedback sanity-checks the quantitative answer once a quarter or so. If your numbers say Instagram stories drives most conversions but your audience consistently mentions reels, look harder at why.
The simplest workable system
For most affiliate creators, the practical setup looks like:
- One bio link per platform (or one bio link that auto-detects the source platform and tags UTMs accordingly).
- A bio platform that shows per-link click counts on free; weekly review on Sunday evening.
- A spreadsheet (or notes file) with one row per active code: code name, brand, start date, expected end date, clicks this week, redemptions last month.
- One audience-side question per quarter (in stories or a newsletter) asking where people found a specific code.
Nothing fancy. The discipline is what matters: writing the data down somewhere weekly is the difference between gut feel and actual signal.
The thing not to do
Don't add a different code per platform per campaign. The idea sounds clever (SAVENAME-IG vs SAVENAME-TT vs SAVENAME-YT attributes the redemption perfectly), but in practice creators forget which is which, audiences see the inconsistency, and the brand's dashboard fragments your numbers into uselessly small buckets. One stable code per campaign + per-link click data on your side gives you 90% of the attribution at 10% of the complexity.
Putting it together
Code attribution isn't a single number you can pull from one dashboard. It's the overlap of three or four signals that each tell you a different slice. The creators who do this well check click counts weekly, redemptions monthly, and spot-check with audience feedback when something looks off. The creators who don't mostly find out after the campaign whether it worked, by which point the answer is already historical.