Meta said the escapes were losing. First-party data said 2.3×.
40 days, 31,382 randomized visitors: +83% checkout conversion and +88% revenue per visitor for escaped traffic — while Ads Manager was simultaneously reporting escaped conversions as a ROAS problem.
The split itself was unambiguous: escaped visitors converted at 2.16% vs 1.18% for control (+83%), and generated $1.00 per visitor vs $0.53 (+88%) over 40 days.
The interesting part was the disagreement with Ads Manager, which showed campaigns rich in escaped conversions underperforming. The cause is structural: iOS strips the fbclid click ID on the Instagram → Safari handoff, so a large share of escaped purchases are invisible to Meta's pixel and get misfiled as direct traffic. On paid-landed sessions, first-party data showed escaped visitors out-purchasing control 2.3× — revenue Meta was attributing to nobody.
EscapeHatch now re-carries the click ID through the escape and restores it before the Meta pixel initializes, closing most of that gap. The takeaway for any brand comparing buckets in Ads Manager: the platform undercounts exactly the visitors who are buying the most.
After the test: The measured window closed in early August 2026 with the result above.
| Arm | Visitors | Orders | CVR | Rev / visitor |
|---|---|---|---|---|
| A · escaped | 16,695 | 360 | 2.16% | $1.00 |
| B · control | 14,687 | 173 | 1.18% | $0.53 |
Brand anonymized. The 2.3× figure is a 14-day paid-landed snapshot from the attribution investigation (100 purchases / $4,783 escaped vs 48 / $2,040 control); the headline numbers are the full 40-day randomized window.
Run this exact test on your traffic.
One script tag, 60-second install. Randomized 50/50 from the first visitor — in 7–14 days this page is your data.