Seven signals show up before an affiliate program breaks: dead new affiliates, a shrinking share of sellers, a quiet top partner, a conversion rate slide, a traffic spike with no sales to match, a reversal spike, and tracking that stops firing. Catch one of them inside 30 days and you fix a dip. Miss all seven for a full quarter and you’re rebuilding a program instead of running one.

I’ve run affiliate programs from $0 to $1 million a month, and programs doing $15 million to $325 million a year. The programs that fell apart didn’t fall apart overnight. They gave off signals for weeks, sometimes months, before anyone noticed the revenue drop. Below are the seven I watch, the number where I stop assuming it’s noise, and the first move I make on each one.
New affiliates who join but never click
More than 30% of the affiliates who joined in the last 30 days with zero clicks logged is where I start worrying. Below that, some no-click affiliates are normal. Above it, something in your welcome process isn’t landing.
This usually means one of two things. Either the welcome sequence never went out (a broken automation, a segment that didn’t fire), or it went out and the ask inside it was too complicated. An affiliate who joined for a directory listing or a one-time bonus and never got a single clear “click this link, post it here” instruction will sit at zero clicks indefinitely.
The first move: pull every affiliate who joined in the last 30 days with zero clicks, and send them one email with one link and one specific place to post it. Not a re-send of the full welcome sequence. One ask. I cover the reasoning behind getting that first click and first sale in how to get affiliates to their first sale, and the logic starts one step earlier, at the click.
A falling share of affiliates making a sale in the last 90 days
A trailing 90-day sale rate under 8%, or a drop of more than 3 percentage points from one quarter to the next, tells me the program is cooling even when total revenue still looks fine. This is a different number from activation rate. Activation counts clicks. This one counts affiliates who converted a sale in the last 90 days.
Revenue hides this problem for a while. A handful of affiliates can carry the total number up even as the percentage of affiliates selling keeps shrinking. By the time revenue notices, the decline has usually been running for a quarter or two already.
The first move: pull the raw percentage today, then pull the same 90-day window from last quarter. If it’s flat or up, you’re fine. If it dropped more than 3 points, something in recruiting, onboarding, or promotion support broke, and a signal check alone won’t tell you which.
If your 90-day sale rate has been sliding for more than one quarter, a signal check by itself won’t explain why. My affiliate program audit walks through every number behind the drop instead of confirming it happened.
A top affiliate going quiet
30 days of zero clicks or sales from an affiliate who normally drives outsized revenue is the threshold. In the programs I’ve run, the top 5 to 10% of affiliates typically generate 80 to 90% of affiliate revenue. That means in a program with 20 or 30 active partners, two or three of them are carrying most of the load. When one of those two or three goes quiet, you didn’t lose one data point. You lost a meaningful piece of the channel.
This one usually means something happened outside your program entirely. The affiliate got busy, changed focus, had a bad experience with a customer of yours, or found a competing offer that pays better. Every one of those has a different fix, and none of them get fixed by a generic email blast.
The first move: reach out personally. Do NOT run this through your email platform’s generic re-engagement automation. Ask what changed. How to reactivate dormant affiliates and how to motivate affiliates who signed up but stopped promoting both dig into the reactivation side of this once you know why they stopped.
A quiet top affiliate is one of the mistakes that can wreck a program if you let it sit too long. My Top 20 Affiliate Program Mistakes report covers this one and 19 others, the same mistakes that nearly sank my program before I built it past $12.6 million in two years.
A conversion rate drop
A relative drop of 20% or more from your trailing baseline, held for two straight weeks and not one, is the number that gets my attention. One bad week is noise. Two in a row is a pattern.
This usually points at something on your side, not the affiliate’s. A landing page changed. An offer got swapped, or a price went up without warning the affiliates sending traffic to it. Checkout added a step. Or the traffic mix changed because a new affiliate is sending a colder audience than your usual partners.
The first move: check whether the drop is program-wide or isolated to one or two affiliates. Program-wide points at your page or offer. Isolated points at the traffic source. What is a good affiliate program conversion rate breaks down what normal looks like by traffic type, which makes the comparison easier.
A traffic spike with no matching sales
Clicks up 25% or more month over month while sales stay flat or fall is the split that should stop you. More traffic with the same or fewer sales isn’t growth. It’s a signal.
The two most common causes are bot traffic and a new affiliate running paid ads to an audience with no reason to buy. Both inflate your click count without adding a single real prospect. A smaller possibility: an affiliate is sending traffic somewhere your customers already saw the same offer, so the clicks are real but the buyers already converted somewhere else.
The first move: check where the spike came from. If it’s concentrated in one or two affiliates or one traffic source, isolate it before you spend a dollar assuming your funnel broke. Affiliate fraud prevention tools covers how to tell real traffic from fake traffic without punishing the affiliates who are converting.
A chargeback or refund jump
A reversal rate above 8 to 10%, roughly double what I consider a healthy ceiling, or a reversal rate that doubles from one pay period to the next, both count as a jump worth stopping for.
A jump almost always means one of three things: a bad batch of buyers came through a specific promotion, coupon abuse or fraud hit a specific affiliate’s traffic, or the product itself has a problem that’s generating more refund requests than usual. Each of those needs a different fix, and lumping them together wastes time.
The first move: pull the reversal rate by affiliate, not only program-wide. If every affiliate reversed at roughly the same higher rate, the product is the problem. If it’s concentrated in one or two affiliates, that’s a traffic or fraud problem specific to them.
Tracking that silently stops firing
Any full day with zero clicks or conversions logged program-wide, not from one affiliate but across the entire program, means your tracking broke, not your affiliates.
This is the signal people miss longest because nothing about it looks alarming at first. No error message pops up. No affiliate emails to complain. The dashboard shows less activity than usual, and it’s easy to assume traffic is slow this week instead of checking whether the pixel is even firing. A plugin update, a theme change, a CDN switch, or a script blocked by a new security setting can all kill tracking without any warning.
The first move: check today, not at the end of the week. How affiliate tracking works covers the full chain from click to recorded commission, which is the fastest way to spot exactly where it broke.
I own AffiliateHQ, so take this as a disclosed recommendation, not a neutral one. Its AffiliateHQ dashboard runs 22 alert types that watch for signals like this around the clock, so a dead pixel or a quiet affiliate gets caught the same day instead of three weeks later during a monthly reconciliation.
The weekly check that catches all seven
Fifteen minutes, same day every week, checking seven numbers, is enough to catch every signal above before it turns into a real problem. Here’s the order I check them in:
- New affiliates from the last 30 days: any at zero clicks?
- 90-day sale rate: up, flat, or down from last week’s check?
- Your top 3 to 5 affiliates by revenue: any gone quiet in the last week?
- Conversion rate: trending down two weeks running, or one bad week?
- Click-to-sale ratio: any traffic spike without matching sales?
- Reversal rate: normal range, or a jump?
- Yesterday’s clicks and conversions: did tracking fire?
Most weeks, all seven come back clean and the check takes five minutes. That’s fine. The point isn’t to find a problem every week. It’s to catch the one week something’s wrong before it becomes a quarter of lost revenue. Affiliate program KPIs covers the full set of numbers worth tracking beyond these seven, if you want to build this into a fuller dashboard.
Running this check alone gets harder as your program grows past a few dozen affiliates. Two ways to get help with it: software that watches these numbers for you, or someone else looking at your numbers with you.
Running these checks alone gets harder the bigger your program gets. My Your Affiliate Launch Coach call is free, 20 minutes, and gets you a second set of eyes on your numbers plus a plan for the next 30 to 60 days.
Pick one signal from this list you haven’t checked in the last 30 days, and check it today. Then set the weekly routine. That’s the whole system: seven numbers, one day a week, no more guessing about whether the program’s healthy or quiet. If you want the fuller system behind these numbers, The Book on Affiliate Management covers it end to end.
