We’re scaling spend and need cleaner attribution to decide where the next $10k/day goes. In a 14-day geo holdout, Meta’s 7-day click ROAS was 3.1 while GA4 showed 2.2 and blended MER was 2.8; funnel CR moved from 3.4% to 3.9% after PDP tweaks, but channel credit is murky. For those who’ve run both Northbeam and Triple Whale, which gave you more trustworthy path-to-purchase insights for budget shifts and CRO prioritization?
Ran both at ~$80k/day; @Northbeam’s Path Explorer + probabilistic model lined up with 14‑day geo holdouts within about 10–15%, while @TripleWhale nailed MER but over‑credited Meta on 1–7 day clickers. Standardize UTMs and turn on server‑side events, then sanity‑check Northbeam’s “Blended” model against the holdout; if your team lives in Shopify, TW’s dashboards are faster to act on. It felt more like following breadcrumbs than confetti.
Quick tip: with a 14‑day geo holdout, we got more trustworthy paths from @Northbeam once we turned off view‑through and forced 7‑day click, then excluded brand search + email from the path report — Meta’s share fell from about 65% to 48% and it lined up within about 8–12%. @TripleWhale was great for tracking MER, but it over‑credited retargeting after the PDP changes bumped CR to 3.9% (); re‑run path‑to‑purchase only on post‑PDP‑change dates before moving the next $10k/day.
And northbeam got a lot cleaner for us after we set a source priority so fbclid/gclid clicks outrank session referrer; before that, retargeting + brand search soaked up too much credit post‑PDP. @s_gregory That tweak put NB within about 8% of our holdout, while Triple Whale still leaned Meta a bit — worth a quick test?