When a contractor says a paid lead vendor is bad, the vendor may be bad. But that should not be the first assumption. Before canceling a source, the business needs a clean record of what happened after each lead arrived.
Lead source cleanup is the step between frustration and a useful decision. It does not defend the vendor. It does not blame the office team. It turns a messy argument into a short evidence list: which leads were real, which were duplicates, which were outside the service area, which were contacted, which were delayed, and which had no next action at all.
The first check is arrival evidence. Did the lead arrive by email, CRM, text, call, portal, or webhook? Was the timestamp visible? Did the team know where to look? A lead that sat in a portal for four hours is different from a lead that reached the inbox immediately. Without that distinction, the vendor and the contractor can both talk past the real issue.
The second check is fit evidence. A bad-fit lead should have a reason that a manager can read later. Wrong service, wrong city, no budget, duplicate request, tenant without authority, job too small, emergency already solved, and no valid contact are different problems. If the team only writes bad lead, the next budget decision has no teeth.
The third check is response evidence. A real lead can still be lost if the first call was late, the voicemail was generic, the text did not mention the requested service, or the estimate follow-up never happened. That does not mean the source is good. It means the business should not decide source quality from lead count alone.
A useful review uses a small sample first. Take 20 to 25 recent leads from the source, remove private customer details that are not needed, and mark five outcomes: ready, hold, duplicate, do not contact, and missing context. The missing-context group is usually the most revealing. It shows where the team cannot prove what happened.
If most leads are wrong-fit with clear arrival and response evidence, cutting the vendor may make sense. If many leads are reachable but unowned, delayed, or mixed with duplicates, the better first move may be a handoff cleanup. If the sample is too thin to tell, the business should fix its tracking before making a larger budget decision.
The point is not to keep spending. The point is to stop guessing. A contractor should cut a paid lead vendor from evidence, not from a week of irritation and a CRM full of half-notes.
That evidence can also make a vendor conversation cleaner. Instead of saying the leads are terrible, the contractor can say which records were outside service area, which were duplicates, which were unreachable after documented attempts, and which were lost inside the handoff. That is a stronger business conversation.