Eleven test visitors, each run with and without a test flag, and the same decision every time

    Who this was forA B2B company whose website visitor campaign hands each identified visitor to four other tools.

    A dry run that exercises the real path

    All case studies

    The drawing

    noyesVisitor eventSame four filters,same decisionTest flag set?Sequencer, enrichment,LinkedIn, CRMAll four hand offsskippedLog line says whateach would have done

    The problem

    The visitor campaign runs with nobody touching it. That's the point. It's also the risk when you want to test a change.

    A test visitor that reaches the sequencer becomes an email to a real inbox. A test visitor that reaches the CRM writes a note on a real contact. And a simulation of the whole path tests a copy, not the thing that runs.

    What we built

    A test flag on the incoming event. With the flag set, the event runs the real path: the same four filters, the same decision, and a real visitor row, marked as a test. Then all four outbound hand offs stop: the sequencer, the enrichment table, the LinkedIn tool, and the CRM. Each one logs a skip in place of its result.

    A skip alone doesn't say what was held back. So the log line also predicts what each hand off would have done, worked out by the same branches in the same order. A real run's log line didn't change by a byte.

    The flag is set when a row is created and never on a repeat visit. A test aimed at a real person's profile can't relabel that person as a test. Cleanup is one delete on the flag.

    What it produced

    • 28 new tests. They record every outbound call the handler would make, not which helper it called.
    • 11 visitor events, each run twice, with and without the flag. The decision, the reason, and whether a row was written matched on all 11.
    • 0 filters behave differently in test mode.
    • The suite went from 662 tests to 690, with 0 failing.
    • Every hand off answers success in the test harness on purpose, so a hand off that wasn't really skipped shows up as a difference, not as a quiet failure.

    Your scenario

    If your automation pushes leads into Instantly or HeyReach, give it a test mode that runs the real path and stops at the hand off. Our Sequencer Lead Push on Apify works that way: a dry run returns the exact payload it would send, and sends nothing. Send us your list of hand offs and we'll tell you where we'd put the stop.

    Related tools: Instantly, HeyReach, Apify, Sequencer Lead Push on the Apify Store, Sequencer Lead Push

    Some of the links on this page are affiliate links.

    Client names are always kept confidential at MMG, so they have been removed from this case study.

    Published August 11, 2026

    Last updated August 29, 2026