Fake Leads vs Bad Leads: How to Tell Automation From Poor Lead Quality
A lead can be genuine and still be unsuitable for your business. Separate qualification from identity validity and automation evidence. Consistent outcome categories make source comparisons more useful and prevent a sales rejection from silently becoming a claim of fraud.
Quick answer
A lead can be genuine and still be unsuitable for your business. Separate qualification from identity validity and automation evidence. Consistent outcome categories make source comparisons more useful and prevent a sales rejection from silently becoming a claim of fraud.
For marketing, analytics, media buying and security teams reviewing paid traffic quality. This guide is technical and operational information, not a legal finding about a specific source.
Fake does not mean unqualified
An enquiry outside the service area, budget or eligibility rules can still reflect real interest. Conversely, a plausible profile may contain stolen or invented contact details. Define why a lead is rejected. The commercial qualification decision and the technical question of fabrication should remain separately reviewable.
Four useful categories
These are operational categories for a lead review, not a universal industry taxonomy. Add pending or inconclusive when follow-up or evidence is insufficient. Do not force those records into the four completed outcomes.
Qualified
The enquiry is confirmed and meets the business criteria agreed for this campaign. Record those criteria and who accepted it. Qualification is a stage in the funnel, not proof of final revenue or valid advertising billing.
Genuine but unqualified
A real enquiry fails a suitability rule such as location, budget or required service. The appropriate response may be offer or targeting review rather than bot blocking. Keep the reason so the source can be evaluated fairly.
Invalid identity/contact
Contact details fail documented checks, conflict or belong to someone who denies the enquiry. Record the exact result, distinguishing a typo or temporary failure from verified misuse. Invalid data alone does not establish automation.
Automated/fabricated
Supporting request, session or submission evidence indicates automation or fabrication. Record those findings separately where possible: not all automation is abusive, and a human can submit fabricated information. Neither classification establishes the actor's identity by itself.
Why CPL hides the difference
Cost per recorded lead mixes outcomes if every submission is counted together. Report the denominator and cohort age beside the cost. A source with cheap forms and few qualified enquiries may consume more sales capacity than one with fewer, better contacts. The comparison must also account for sales process and attribution gaps.
Calculate qualified-lead cost and sales workload without double counting
Contactability
Define a fair attempt policy and record delivery, response and confirmation outcomes. One unanswered call is not proof that a person is fake. Separate invalid numbers, undeliverable mail, denied enquiries and ordinary non-response. Check whether business hours or delayed follow-up explain source differences.
Qualification
Agree criteria before reviewing traffic. Record missing information as pending rather than an automatic failure. Use the same rule across compared sources, or explicitly explain why different offers need different criteria. Changes to the rule create a measurement break that should be visible in the report.
Sales acceptance
A qualified enquiry can still be rejected by sales because of capacity, duplicate ownership or process errors. Record acceptance and its reason separately. A handoff failure should not be attributed to ad fraud without evidence about the original enquiry.
Revenue
Follow accepted opportunities to the relevant business outcome where data permits. Account for sales-cycle delay, cancellations and duplicates. Lack of a sale does not retrospectively make a lead fake. A source comparison should show which cohorts have matured and which remain open.
How classification improves bidding signals
A consistent outcome definition can support more meaningful conversion feedback. Google offers qualified and converted lead categories, but matching and configuration are advertiser responsibilities. Review the outcome rule and data path before changing goals; this guide does not imply AdFraud.pl uploads CRM statuses or automatically adjusts bidding.
Connect lead classifications to the conversion goals used for biddingAssess traffic against confirmed business outcomes
Related reading
Sources and further verification
Platform rules, standards and detection practices change. Check the current primary source and your own measurement scope before applying a control.