Why your affiliate network and your tracker report different conversions
Why your affiliate network and your tracker report different conversions
A conversion discrepancy between your affiliate network and your tracker is almost never a mystery. It comes down to a short list of mechanical causes: the network counts a conversion your tracker never received (postback failed), counts it twice (no dedup), counts it on a different day (timezone), or counts a different thing entirely (lead vs sale, approved vs pending). Before you assume the network is shaving you or your tracker is broken, walk the chain. In most cases you can reconcile the two numbers to within a handful of events — and explain the handful that remain.
This guide covers the causes in the order they're worth checking, with the specific settings and test steps that isolate each one. If you only want the fix and not the reasoning, start with the step-by-step troubleshooting guide for affiliate conversions not tracking and come back here when you need to understand the numbers.
What causes a conversion discrepancy between an affiliate network and a tracker?
There are seven causes that account for nearly every mismatch. Ranked by how often they're the culprit:
- The postback never arrived. The network recorded the conversion; your tracker didn't. This is the single most common cause of a tracker showing fewer conversions than the network.
- The postback fired but was rejected. Wrong click ID, wrong parameter name, expired click, or a duplicate that got dropped — the request hit your tracker and was thrown away.
- Timezone and reporting-window differences. The network reports on its clock and its day boundary; your tracker reports on yours. Conversions near midnight land on different dates.
- Deduplication differences. One side dedupes by order ID, the other doesn't — or dedupes on a different key.
- Conversion type mismatch. The network counts a lead, a sale, and an upsell as three events; your tracker has one "conversion" and counts one. Or the reverse.
- Status filtering. The network's dashboard shows approved conversions; your tracker shows everything it received, including later-reversed ones. A reversed sale stays in your tracker unless you update it.
- Attribution lag and delayed postbacks. Some verticals (finance, lead gen with manual review) post back hours or days after the click. Your tracker's click may have aged out, or you're comparing a live number to a settled one.
Notice what's not on the list: shaving. It happens, but it's rare relative to the boring mechanical stuff above. Prove the mechanics first. If the chain is provably clean — postbacks arriving, click IDs matching, dedup configured — and the network still shows materially more conversions, then you have a conversation to have.
Step 1: Establish which side is wrong before you touch anything
You can't fix a discrepancy until you know which number is true. Pull a single conversion from the network's report and trace it end to end.
- Pick a specific conversion in the network's report — one with a timestamp and, ideally, the click ID or sub-ID the network captured.
- Search that click ID in your tracker's conversion log.
- Three outcomes:
- Found, matching payout and timestamp — both sides agree on this one. The discrepancy is elsewhere; go to Step 3.
- Not found — the postback didn't arrive or was rejected. Go to Step 2.
- Found but different payout or timestamp — a value-passing or timezone problem. Skip to Step 4.
Do this for five conversions across different days and geos. If all five are missing from the tracker, you have a systemic postback failure. If two of five are missing, you have an intermittent one — usually a timeout, a status-code rejection, or a click that aged out.
Step 2: Check the click ID is actually surviving the round trip
The postback is a request the network sends to your tracker containing the click ID it stored when the user clicked. If that click ID is missing, truncated, or renamed, your tracker can't match the conversion to a click and drops it.
The chain looks like this:
- Your tracker generates a click ID and appends it to your affiliate link as a sub-ID parameter (e.g.
sub1={clickid}). - The network receives the click, stores the sub-ID value against that click.
- On conversion, the network fires your postback URL with the stored value substituted in:
https://track.yourdomain.com/postback?cid={sub1}&payout={payout}. - Your tracker receives
cid, looks it up, and records the conversion against that click.
Every one of those four steps can break. The most common failures:
- Parameter name mismatch. The network sends
subid; your postback expectscid. The value arrives but your tracker doesn't read it. Check the exact macro name in the network's postback documentation against your postback URL. - Macro not substituted. The network sends the literal string
{sub1}instead of the value. This usually means the macro isn't supported on that network's postback, or you used the wrong token. - Click ID truncated. Some networks cap sub-ID length. If your click IDs are long, they get cut and no longer match.
- URL encoding. Ampersands, plus signs, or spaces in the click ID break the query string. Confirm the value is URL-encoded.
If you're not sure your tracker is even generating and appending a click ID correctly, read what a click ID is and why conversions go missing without one first. On Adtraxo, sub-IDs sub1–sub5 are Pro features — the tracker appends {clickid} into whichever sub-ID slot you assign, and the postback URL it generates for you already expects the matching parameter. If you're on Free, postback conversions aren't available, so any network-side conversion will show as a discrepancy by design; that's a plan limitation, not a bug.
Step 3: Confirm the postback is firing and being accepted
"Not found in the tracker" splits into two very different problems: the postback never left the network, or it arrived and your tracker rejected it. You need to see the actual request.
- Fire a test conversion. Most networks have a test-postback button in the affiliate link settings. Use it and watch your tracker's conversion log in real time.
- If nothing appears, the request isn't reaching you. Check the postback URL character by character: protocol, domain, path, parameter names. A single typo in a parameter name is enough to silently drop a conversion.
- If the request arrives but no conversion is recorded, your tracker is rejecting it. Common reasons: the click ID doesn't exist (expired click or wrong ID), the click ID is a duplicate (dedup working correctly), or the request used the wrong HTTP method.
- Look at your tracker's raw request log if it has one. Adtraxo logs inbound postback requests with their status, so you can see whether a request arrived, what parameters it carried, and whether it matched a click.
If you're getting nothing at all, work through postback not firing: the checks to run, in order. And before you spend another dollar on traffic, run how to test a postback URL before you send paid traffic — it takes ten minutes and catches the majority of these failures.
One more thing worth confirming at this stage: that you're using a server-to-server postback and not relying on a client-side pixel. Postbacks are far more reliable because they don't depend on the user's browser, ad blocker, or cookie state. If you're weighing the two, postback URL vs tracking pixel covers when each is appropriate. Adtraxo's S2S postback conversions are a Pro feature, and they're the right default for reconciliation because the request is server-generated and logged.
Step 4: Reconcile timezones, conversion types, and payout values
If the missing conversions are few and the totals are close but not equal, you're probably looking at one of these three.
Timezones and day boundaries
Your network might report in UTC, your tracker in your account timezone, and your traffic source in yet another. A conversion at 23:50 UTC on the 14th is the 15th in much of Europe. Over a month this shifts a handful of conversions across the boundary and makes daily numbers look wrong while monthly totals match closely. Fix: set your tracker's reporting timezone to match the network's, or compare on monthly totals rather than daily until you do.
Conversion types and event counts
Networks often distinguish lead, sale, upsell, and rebill as separate conversion types. If your tracker has one conversion event, a sale-plus-upsell shows as one conversion in your tracker and two in the network. The reverse happens if you've set up a single postback for multiple event types and your tracker dedupes them.
Before you compare totals, confirm both sides are counting the same thing. If you run lead-gen offers, track leads and sales as distinct events so the comparison is apples to apples — see how to track leads and sales as separate conversion events.
Payout values
Count discrepancies are one thing; value discrepancies are another, and value errors quietly wreck your ROI math. If the network reports $42 in revenue and your tracker reports $0 or $1, the payout macro isn't being passed or isn't being read. Check that your postback URL includes the payout parameter (e.g. &payout={payout}) and that your tracker is configured to read it as revenue, not as a fixed default. How to pass payout values in a postback so your ROI numbers are real walks through the parameter setup.
Step 5: Rule out duplicates and reversals
Two smaller causes that produce a persistent, frustrating gap:
- Duplicates. If the network fires the postback twice (retry, double callback, user refreshing a thank-you page), your tracker records one and the network records one — fine. But if your tracker doesn't dedupe, you'll show more conversions than the network. Check whether your tracker dedupes on a unique order/transaction ID and whether the network passes one.
- Reversals and status changes. Networks reverse conversions for fraud, chargebacks, or failed manual review. The network's dashboard may show only approved conversions while your tracker still holds the original. If you're comparing "approved" on the network to "all received" in your tracker, the gap is expected. Reconcile on the same status.
A clean setup passes a unique transaction ID in the postback and dedupes on it. If your network supports that parameter, use it.
Step 6: Check the traffic side — did the click even reach the network?
Sometimes the discrepancy isn't in the conversion at all; it's in the click count. If your tracker records 1,000 clicks and the network records 940, you have a click discrepancy, and conversions will naturally differ too. Causes: redirect chains dropping users, a landing page that fails to load for some visitors, bot traffic the network filters out, or a tracker click that never resulted in a redirect to the offer.
This is where a one-line arrival snippet earns its keep. Any Adtraxo plan lets you add a snippet to your landing page that records which clicks actually loaded the page in a real browser. That separates "click recorded by the tracker" from "human who saw the page" — and if the network's click count is lower than your tracker's, the snippet tells you whether the difference is non-arriving clicks or network-side filtering. It's the fastest way to prove the traffic side is clean before you blame conversions.
Also confirm your sub-ID or UTM setup is passing the right values. Mixing the two, or passing a UTM where the network expects a sub-ID, means the network stores nothing usable and can't post back against the click. UTM parameters vs sub-IDs explains which to use on affiliate links.
How to keep the two numbers reconciled going forward
Stop comparing daily totals and start reconciling on a fixed schedule with a fixed method:
- Pick one timezone and set both the network report and your tracker to it.
- Compare the same status — approved to approved, or all to all.
- Reconcile weekly on settled data, not live numbers, so delayed postbacks have landed.
- Keep a written tolerance. A small gap from timezone boundaries and reversals is normal. A gap that grows week over week is a postback failure you haven't found yet.
- Log every change to postback URLs and parameters with a date, so when a discrepancy appears you know what changed.
If you run multiple traffic sources — PropellerAds, EvaDav, RichAds, Taboola, MGID, solo ads — reconcile per source. A discrepancy isolated to one source usually points to that source's click handling or a campaign-specific postback override, not your tracker.
And if you're running email traffic, the same click-ID discipline applies to identifying which message drove the sale: how to track which email in your follow-up sequence drives affiliate sales.
The tools don't have to match to the conversion. They have to match closely enough that you can explain every missing one. When you can, you know your tracking is sound and your optimization decisions are based on real numbers.
Frequently asked questions
Why does my affiliate network show more conversions than my tracker?
Most often the postback failed to arrive or was rejected — a wrong parameter name, a missing click ID, or an expired click. Less often it's a timezone difference, a status filter (the network counts approved, your tracker counts everything), or the network counting lead and sale as two events where your tracker counts one. Trace a single missing conversion end to end to isolate which it is.
Is a small conversion discrepancy normal?
Yes. Timezone day boundaries, delayed postbacks, and reversals will always leave a small gap between two systems that count on different clocks and statuses. What matters is that the gap is small, stable, and explainable. A gap that grows week over week is a postback failure you haven't found yet.
How do I test whether my postback is working?
Use the network's test-postback button and watch your tracker's conversion log in real time. If nothing appears, the request isn't reaching you — check the URL character by character. If it arrives but no conversion is recorded, your tracker is rejecting it, usually because the click ID doesn't match or is a duplicate. Adtraxo's inbound request log shows whether each postback arrived and what it carried.
Why is my tracker showing the wrong conversion value?
The payout macro isn't being passed or isn't being read. Confirm your postback URL includes the payout parameter (e.g. &payout={payout}) and that your tracker is configured to read it as revenue rather than a fixed default. Without it, your ROI math is built on a number that isn't
Track your affiliate campaigns for free
Clicks, conversions, fraud detection — all in one tracker.
Get started free →Not ready for an account? Get the Tracking Domination Playbook — free, just your email.