Why leads get returned
This guide explains what a returned lead is, the reasons buyers in the lead market most often give for returns, and how to trace returns back to the traffic that causes them. By the end you will be able to sort your returns by sub ID, match each pattern to a likely cause, and make changes to your ads, targeting and sources that cut returns without cutting sold leads.
Published 5 October 2026. How we write and check guides: editorial standards.
Already generating leads on your own forms? Sell your leads instead.
What a returned lead is
A returned lead is a lead that sold to a buyer and was later sent back because the buyer found a problem with it. Across the lead market, buyers agree return rules with the networks they buy from. Those rules set which reasons count and how long a buyer has to raise them. When a return is accepted, the sale is reversed.
For you as a publisher, this matters because you are paid rev share on sold leads, and returns are deducted under the return terms agreed during onboarding. A lead that sells and then comes back does not add to your payout. Unsold leads are not paid either, so the leads that count are the ones that sell and stay sold. See how payouts work for the full picture.
Returns are also a signal. A buyer who keeps returning leads from one source will often bid less for that kind of lead, or stop bidding. In a ping/post market, that lowers what your future leads sell for. Fixing the cause of returns protects both today's payout and tomorrow's demand.
This page describes common return reasons across the lead market in general. It is not Summit's return policy. Your return terms are agreed during onboarding, so ask then if anything about them is unclear.
Common return reasons in the market
Buyers describe return reasons in different words, but most fall into a small set of groups. The table shows each group, what usually causes it on the traffic side, and where to look first.
| Return reason | What usually causes it | Where to look first |
|---|---|---|
| Bad contact details | Wrong or disconnected phone numbers, typos, fake names, mailboxes that bounce | Sources and placements with accidental clicks, or ads that offer something for free |
| Did not request contact | The person does not remember filling in a form, or says they never did | Ads that do not say what happens after the click, and incentive-style hooks |
| Does not match the offer | An MVA lead with no accident or no injury; a home improvement lead who rents or wants a different service | Ad angle and audience. Vague copy draws people outside the offer |
| Already has help | An injured person who already has a lawyer, or a homeowner who already signed with a contractor | Ads that reach people late, after they have acted |
| Outside the buyer's area | The person lives outside the buyer's service area or state | Location targeting that is too loose, or ads served outside your chosen states |
| Duplicate | The same person submitted more than once, or was already sent to the buyer | Retargeting the same people repeatedly, or running the same audience across many ad sets |
| Suspected fraud | Form fills that look automated or come from the same device or network many times | Unknown traffic partners, bought traffic you cannot see, and sudden spikes from one placement |
The details buyers look at differ by vertical. For MVA, see what buyers check on the car accident offers page. For home improvement, see roofing offers and the other category pages. The exact qualifying criteria are confirmed with each publisher during onboarding.
Step 1: Sort returns by sub ID
Leads are reported by tracking link, and your sub IDs break that report down by campaign, ad set and ad. Without sub IDs, you can see that returns happened but not where they came from. If your links do not carry sub IDs yet, start with tracking links and sub IDs explained before anything else.
- Pull the leads for a set period, with their status (sold, unsold, returned) and their sub IDs.
- Group the leads by your top sub ID level, usually campaign or traffic source.
- For each group, count sold leads and returned leads. Look at the share of sold leads that came back, not the raw number.
- Take the group with the most returns relative to its sold leads and break it down one level further, to ad set and then ad.
- Stop when you find the smallest unit that explains most of the returns. That is the thing to change.
Look at a long enough period that a few returns do not swing the picture. Returns can arrive some time after the sale, so the most recent days will show fewer returns than they will end up with. Compare periods that have had the same time to mature.
Step 2: Match the pattern to a cause
Once you know where returns come from, the pattern usually points to the cause. Use the return reasons you are given, if any, together with these questions:
- Is it one ad or the whole ad set? One ad points to copy or creative. A whole ad set points to the audience, placement or states.
- Is it one source? If returns cluster on one network or one placement type, the problem is the traffic, not the offer.
- Did it start on a date? A sudden rise often follows a change: a new ad, a new placement, a budget jump, or a new partner sending traffic.
- Are the returned leads alike? Many returns from people outside your states point to targeting. Many returns for "did not request" point to the ad promise.
Step 3: Fix the cause
Bad contact details
Remove placements that produce accidental clicks, such as in-app placements where taps are easy to make by mistake. Rewrite any ad that offers something free or instant, because those draw people who type anything to get through the form. Keep the lander as provided; consent language on Summit landers is not changed or removed.
Did not request contact
Say in the ad what happens next: a short form, then a call or message about options. A person who knows a call is coming is far less likely to say they never asked for one. Read TCPA basics for lead gen affiliates for why clear expectations and consent matter.
Does not match the offer, or already has help
Make the angle specific. Name the event and the time frame for MVA ("hurt in a car accident recently?"). Name the job and ownership for home improvement ("own your home and need a new roof?"). Run ads earlier in the person's decision, which on search means question and research terms as well as late-stage terms.
Outside the area, duplicates and fraud
Tighten location targeting to your chosen states (see geo-targeting by state). Cap how often the same people see retargeting ads. Stop any traffic you buy from others until you can see its source by sub ID, and read bot and fraud traffic: how networks detect it.
Running traffic that stays sold? Apply with your sources, states and expected volume.
Apply as a publisherPrevention checklist
Run through this list before you launch a new campaign, and again whenever returns rise:
- Every ad links to your tracking link with sub IDs for campaign, ad set and ad.
- Every ad says what the offer is and what happens after the form.
- Every ad gives the reader one reason to click: help with the accident or the home project itself.
- Location targeting is set to the states you chose, with people in those states, not just interested in them.
- Accidental-click placements are off or tested on their own in a separate ad set.
- Retargeting audiences exclude people who already submitted.
- Any traffic from a third party carries its own sub ID so you can cut it off alone.
- The lander and its consent language are used as provided. See traffic guidelines.
Common mistakes
- Reading the latest days as final. Returns arrive after sales, so recent results look better than they will end up.
- Cutting the source with the most returns by count. It may also have the most sold leads. Compare returns to sold leads in each group.
- Fixing returns by narrowing everything at once. Change one thing per test so you know what worked.
- Ignoring the ad promise. Copy that overpromises brings in people who are surprised by the follow-up call.
- Buying traffic you cannot see. If a partner sends traffic under your link with no sub ID of its own, you cannot remove the bad part.
- Treating returns as a dispute to win. The payout comes back fastest when you remove the cause, not when you argue each lead.
Frequently asked questions
Are returned leads paid?
No. You are paid rev share on sold leads, and returns are deducted under the return terms agreed during onboarding. See how payouts work.
Is the list on this page Summit's return policy?
No. It describes common return reasons across the lead market. Your return terms are agreed during onboarding.
How do I know which ad caused a return?
Leads are reported by tracking link. If every ad carries its own sub ID, you can sort returned leads by sub ID and find the campaign, ad set or ad behind them.
Can good traffic still get returns?
Yes. Some returns happen on any source, for example when a person changes their mind or gives a wrong number by mistake. What matters is whether one part of your traffic returns far more than the rest.
Related guides
Get more of your leads to sell in the first place.
Sold, unsold and returned leads by sub ID.
How networks spot traffic that will not stay sold.
Set up the IDs that make returns traceable.
Rev share, unsold leads and returns.
Send traffic that stays sold
Every publisher is reviewed before going live. Email us your sources, states and how you track your traffic.
Apply as a publisherOr write to team@summitleads.ai. We reply by email.
Already generating leads on your own forms? Sell your leads instead.