Why your attribution numbers come from us, not from ad platforms
Ad platforms over-report. Your bank doesn’t. Admaxxer reconciles three independent sources — your real revenue ledger, your first-party pixel, and the ad platforms’ own APIs — and shows you the dedupe-anchored truth. This page is the canonical answer for “why don’t Admaxxer’s numbers match Meta?” and “how do I know your attribution is right?” — written for both ecommerce stores and SaaS subscription businesses.
How we deduplicate attribution truth
Three independent sources answer the same question — “who paid you, and which marketing touch drove the sale?” — from three different vantage points. Each source has blind spots; none of them is individually correct. We reconcile them, show you where they disagree, and pin the headline number to the one source you can verify against your bank statement.
Source 1 — Your revenue ledger
The accounting truth. Shopify Admin for ecom, Stripe (or Paddle, Polar, LemonSqueezy, Dodo) for SaaS. This is the source you reconcile against your bank statement and report to investors. It knows about every paid customer, but doesn’t see marketing context — which ad they clicked, what they read before checkout, whether they came back through email.
Source 2 — Your first-party pixel
The click-side truth. Our pixel fires on your site under your domain, so it sees real visitors across browsers, devices, and privacy modes that platform pixels can’t see. It captures every touchpoint — paid clicks, organic search, direct, referral, email — and stitches them into the path that led to checkout. It can miss revenue from offline checkout or refund chargebacks that happen after the visit.
Source 3 — The ad platforms’ own API
The platform’s self-reported credit. Meta, Google, TikTok, Amazon, Pinterest. Each platform claims credit for conversions its own algorithm believes it caused — including view-through conversions and modeled-conversion estimates that go beyond what was actually clicked. The number is useful for campaign optimization signal but systematically over-reports versus the ledger.
How we combine them
- Pin to the ledger. Top-line revenue, customer count, and average order value anchor to your ledger. If Shopify says you did $48,200 last week, our headline says $48,200. The lens-switching toggle exposes the other sources alongside, but the default is always the number you can audit.
- Stitch the marketing path. The pixel’s session-and-touchpoint graph joins each paid customer to the actual marketing journey that drove them. First-touch, last-touch, linear, time-decay, position-based, and data-driven Markov — choose the model that fits the decision you’re making.
- Reconcile the platform claims. The Reconciliation Panel shows you, per channel, the spread between what Meta or Google reported versus what your pixel saw versus what your ledger recorded. The gap is explained line-by-line — privacy blocks, modeled conversions, refunds, delayed attribution — so you always know the why behind a difference.
What this looks like in your dashboard
The three-source reconciliation works the same way for everyone, but the KPIs, lenses, and example numbers are different for an ecommerce store than they are for a SaaS subscription business. Admaxxer picks the right surface for your workspace automatically — set your business type once during onboarding and every tile speaks your language.
If you sell physical products (ecommerce stores)
Your headline anchors to Shopify Admin. Marketing attribution flows from the pixel’s click-side journey. Platform numbers show what Meta and Google think they caused — almost always more than your bank actually saw.
Worked example — a $79 supplement. Your $79 supplement runs a Meta Advantage+ campaign for the week. Three different numbers show up:
- Meta claims 12 conversions from the campaign (its in-platform conversion column).
- Our pixel attributed 10 Shopify orders to that campaign’s click path.
- Shopify Admin shows 10 paid orders from those customers (the ledger truth).
Admaxxer’s headline: 10 orders, $790 revenue, $79 AOV — the ledger truth. Meta’s 12 stays visible in the Reconciliation Panel so you can see the platform’s optimism, and the spread is explained inline. Primary ecom KPIs: ROAS (revenue divided by spend), AOV (cart truth from Shopify), MER (blended marketing efficiency across all channels). Refund-aware, partial-order-aware, multi-currency-aware (each row carries its own historical exchange rate so cross-currency totals still tie to the ledger penny-for-penny).
If you sell a subscription (SaaS businesses)
Your headline anchors to Stripe (or Paddle, Polar, LemonSqueezy, Dodo). Attribution follows the signup event through trial conversion, expansion, and renewal — across the same three independent sources. The unit is MRR and ARR, not AOV.
Worked example — a $29/month subscription. Your $29/month plan runs a Google search campaign on a high-intent keyword. Three different numbers show up:
- Google claims 9 conversions (its in-platform sign-up event count).
- Our pixel attributed 8 signups to that campaign’s click path.
- Stripe shows 7 paid customers (one trialled but didn’t convert).
Admaxxer’s headline: 7 new customers, $203 new MRR, $2,436 added ARR — the ledger truth. The 8 signups stay visible as trial-stage attribution, the 9 platform claims as the optimization signal, and each lens has its own tooltip explaining the gap. Primary SaaS KPIs: MRR (from Stripe), LTV (3-24 month cohort lens), CAC (per-channel acquisition cost with payback period). Trial-to-paid conversion, expansion MRR, contraction MRR, and churn cohorts are all attributed back to the original acquisition source.
Set your business type once. We never mix ecom KPIs with SaaS KPIs on the same surface. During onboarding you pick your business type (or switch later from Settings) and the entire dashboard adopts the right lenses, the right currency formatting, the right lifecycle stages, and the right comparison benchmarks. If you run both — say, a SaaS company with a one-time setup-fee product — each workspace is independent and you can run one of each.
What the multi-source approach gives you
The standard ad-platform-only attribution model is broken in three well-documented ways. Multi-source reconciliation isn’t a speculative new approach — it’s the same discipline that sophisticated DTC and SaaS finance teams have used for years, now built into the dashboard so you don’t need a data team to run it.
- Three independent vantage points. Ledger + pixel + platform — every paid customer triangulated.
- 100% ledger-anchored. Top-line revenue ties to Shopify or Stripe to the penny.
- Six attribution models. Last-click, first-click, linear, time-decay, position-based, and data-driven Markov.
- You always know the source. Every number on every tile carries a source badge. Hover any tile and you’ll see which lens it came from — pixel, platform, or ledger — plus the timestamp of the last refresh. Numbers with footnotes get a tooltip explaining the dedupe rule applied. No black boxes.
- You can override the default. The default headline anchors to the ledger because that’s the number most teams trust most often, but every tile has a lens toggle. Switch any KPI to pixel-only (the click-side view) or platform-claimed (the optimization-side view) for the question you’re trying to answer.
- Refunds, partials, and renewals. The ledger anchor naturally handles refunds (Shopify), partial refunds (Shopify), failed renewals (Stripe), trial cancellations (Stripe), and dunning recovery (Stripe). What you see in the dashboard always reflects the net-of-refund, net-of-churn truth — never the gross over-report that the platforms surface alone.
- A fourth lens — ask the customer. Post-purchase surveys give you a fourth, qualitative source — what the customer actually says drove them to buy. We calibrate the three quantitative sources against this human ground truth to surface mis-attributed branded search and under-reported referral channels.
Frequently asked questions
The questions support gets most often about Admaxxer’s attribution accuracy. Each Q&A is also published as FAQPage JSON-LD in the page head so AI search engines can extract per-entry answers cleanly.
Why don't my Admaxxer numbers match Meta Ads Manager?
Meta only sees the customers who clicked one of its own ads, and only when it can identify them through its own cookies and signals. Admaxxer sees three things: every customer who actually paid you, every click your pixel tracked, and every conversion the ad platform reported. We reconcile those three sources, dedupe overlaps, and surface the customer count and revenue you can actually trust in the bank. Meta's numbers are an estimate that systematically over-reports — every major attribution study from Northbeam, Triple Whale, and academic researchers shows the same gap, between 15 percent and 60 percent depending on the campaign type. Our headline number is the dedupe-anchored truth.
How do I know Admaxxer's attribution is the right one?
Two anchors. First, your real revenue ledger — Shopify Admin for ecom businesses, Stripe for SaaS — is the universal source of truth. Both are accounting systems you reconcile against your bank statement. We pin our top-line revenue to that ledger and explain every reconciling line item in plain English. Second, every dashboard tile shows which source each number came from with a visible badge. If the badge says Pixel and you want to cross-check against your Stripe invoices, the difference is shown side-by-side so you can see why and decide which lens fits your question. Trust comes from transparency, not from a black-box number.
What if I see different numbers in my workspace than expected?
Open the Sources & Attribution view and use the source toggle. You'll see the same campaign with three columns side-by-side: the platform's own number, our pixel's number, and the revenue ledger number. When the three disagree, the cause is one of a handful of well-understood patterns: an iOS 17 ATT user blocking the platform's view, a delayed conversion the platform attributed later than we saw, an organic search visitor the platform claimed credit for, or a refund that hadn't propagated yet. Each pattern has a tooltip that explains the gap and tells you which lens to trust for which question. If you still can't reconcile a number, support is one click away — every comparison ships with a “flag this” button.
How do you handle iOS 17 ATT and other privacy changes?
ATT, ITP, and the rest of the privacy-bucket changes hide users from the ad platform’s view — but they do not hide users from your own ledger. A customer who clicked an iOS 17 ad with tracking blocked still shows up in Shopify as a paying order, or in Stripe as a paid subscription. Because we reconcile to the ledger, those customers still count for you in Admaxxer. The platform may not be able to attribute the click back to a specific ad, but you still see the revenue and the customer in your numbers. For workspaces that connect both the platform’s own server-side conversion API (CAPI for Meta, Enhanced Conversions for Google) and our pixel, the match rate that recovers the otherwise-hidden conversions is visible on the Tracking Health tile.
Do I need to connect all three sources to get attribution?
No. Each source layer you connect makes the numbers more accurate, but the dashboard works as soon as the pixel is installed. Pixel-only attribution gives you click-side truth and works on day one for both ecom and SaaS. Adding your revenue ledger — Shopify, Stripe, Paddle, Polar, LemonSqueezy, or Dodo — anchors top-line revenue and unlocks the reconciliation lens. Adding the ad platforms' own APIs (Meta, Google, TikTok) closes the third side of the triangle and lights up CAPI match rate, ad-level LTV, and the three-way drill-down. Connect what you have; we’ll show you the gap and let you fill it in your own time.
Related
Sources & Attribution drill-down · Attribution models · Markov data-driven attribution · Tracking Health · Post-purchase surveys · Multi-currency model · Data architecture overview · Revenue tracking model · Documentation home
Questions or feedback: support@admaxxer.com.