Switching

In four weeks
you'll know the truth.

Admetrics runs beside your current setup on the same orders. About an hour to switch on, no developer, and 21 days free to see whether the numbers hold. Nothing gets switched off, and your history is imported from day one.

Plan your switch See pricing

Two measured distributions separating, with 97 percent confidence that one is higher

Three phases. One agreed timeline.

Four weeks, three phases, and a decision that sits with you at the end of it.

Admetrics connecting alongside an existing tracking setup

Week 1. Connect

Admetrics goes in next to what you already run. No code freeze, no migration window, nothing switched off.

Two measured distributions compared against a baseline

Weeks 2–3: Validate

Both systems record the same orders. Results are compared against an independently measured baseline.

Two paths converging on a single decision

Week 3+: Decide

The evidence sits on your own data. Cut over, extend the parallel run, or walk away.

Your history comes with you.

A new tool that starts empty is useless for three months. Yours does not start empty.

Backfilled during setup

Orders, products, costs and platform spend are imported for the past periods your shops and ad accounts can provide. You open your first dashboard with history in it, not a blank slate.

Comparable from day one

Because the past is already there, the first reports mean something: last quarter against this one, last Black Friday against the next. No waiting for data to accumulate.

Switching is a risk problem, not a setup problem.

The install was never the hard part. The risk is committing next quarter's budget to numbers you haven't tested yet. So we don't ask you to. The system you trust today keeps running until the new one has proven itself on your own orders.

The same orders.
Two different records.

During the validation period both systems watch the same traffic and the same checkout. What differs is how much of the journey survives to the report, and that difference is the whole reason to run the test.

A customer journey with missing touchpoints

What your current setup keeps

Browser-side pixels lose sessions to consent, ad blockers, Safari and cross-device hops. The orders still arrive; the touchpoints that earned them go missing, so credit lands on whatever was last visible.

A complete customer journey traced end to end

What Admetrics records

First-party, server-side collection keeps the journey intact and ties every touch back to a named order. You can open any single order and see exactly which channels it passed through.

Three things teams find before the trial is over.

Not predictions. These are the discrepancies that keep coming up once two systems watch the same checkout, and they tend to surface in the first few days.

Orders nobody was crediting.

Consent walls, ad blockers, Safari and cross-device journeys quietly drop conversions from browser-side tracking. Server-side collection picks them back up, and they usually belong to a channel you had already written off.

The same sale, counted three times.

Every platform claims the orders it touched, so the numbers in your spreadsheet add up to more revenue than your shop took. One journey view per order ends the argument about which channel earned it.

Winners that lose money.

A campaign can clear its ROAS target and still be underwater once COGS, shipping, payment fees and returns come off. Contribution margin per campaign tends to reorder the top of the list on day one.

Admetrics helped us overcome scaling challenges and set a new benchmark for paid-ads.
Paul KeilTeamlead Performance Marketing

+ +

Your stack doesn't move.

Your shop, ad accounts, marketplace, email and SMS and warehouse all stay where they are. Admetrics connects through 30+ native integrations and reads alongside your current setup. Nothing gets rebuilt, nothing gets moved off the tools you already pay for, and your live campaigns keep running throughout.

30+ Native integrations

Shopify

Amazon

Shopware

Magento

WooCommerce

JTL

Recharge

Stripe

PayPal

Google Sheets

BigQuery

Amazon S3

Google Drive

Slack

Claude

OpenAI

Gemini

Shopify

Amazon

Shopware

Magento

WooCommerce

JTL

Recharge

Stripe

PayPal

Google Sheets

BigQuery

Amazon S3

Google Drive

Slack

Claude

OpenAI

Gemini

Meta Ads

Google Ads

TikTok Ads

Snapchat

Microsoft Ads

Instagram

Facebook

Criteo

Taboola

Outbrain

Teads

Awin

Tradedoubler

Linkster

Klaviyo

Fairing

KnoCommerce

Zapier

Make

Meta Ads

Google Ads

TikTok Ads

Snapchat

Microsoft Ads

Instagram

Facebook

Criteo

Taboola

Outbrain

Teads

Awin

Tradedoubler

Linkster

Klaviyo

Fairing

KnoCommerce

Zapier

Make

Measurement moves first. Money moves last.

Tracking moves server-side, so the conversions a browser pixel drops get recorded. Profit is calculated below the gross-margin line, on your real costs. Only once both hold up does anything touch a budget.

What you get on the other side.

Three things that are hard to bolt on later and harder to copy: measurement you own, profit calculated on your real costs, and a system that acts on both instead of stopping at the report.

A measured curve continuing past the point where tracking usually stops

Measurement you own

Multi-touch attribution, marketing mix modeling, experiments and first-party server-side tracking resolved into one signal, not a single black-box model you take on faith. Every euro traces back to an individual order.

Contribution profit compounding across stages

True profit, not platform ROAS.

Contribution margin calculated below the gross-margin line on your real COGS, shipping, fees and returns. CM2, POAS and new-customer CAC become the numbers you steer with, per campaign and per product.

Budget shifted between channels with the amount moved labelled

The number moves the budget

Most tools stop at the report. AdPilot turns the measurement into a cross-channel allocation, executes it inside limits you set, and logs every action with its reason, constraint and outcome. Reversible at any point.

Two hours. No engineer. No invoice.

The reason most teams never test their numbers is that testing them sounds like a project. It isn't one. There is nothing to build, nobody to brief and nothing to pay until you have seen the result.

No developer. No code freeze.

Shopify and Shopware connect natively, ad accounts through OAuth. No sprint ticket, no engineering review and no pause on live campaigns. On a supported shop the tracking installs through the app rather than the codebase.

Live in about two hours.

Connections, cost inputs and metric definitions in one sitting, with a success manager alongside you. Past orders and platform spend come in during setup, so you open a dashboard with history already in it rather than waiting a month for data to accumulate.

21 days free, on your own orders.

The full product on your real data, not a sandbox. The gap between what your pixel recorded and what actually happened usually shows up in the first week, well before you decide anything.

You can stop at any point.

There is no lock-in step in the parallel run. Your existing tracking keeps recording throughout, exports stay open, and the data stays yours. If the numbers don't convince you, nothing has been dismantled.

What the switch actually costs you.

Around two hours of setup, a few weeks of running two systems side by side, and one uncomfortable meeting when the conversion counts disagree. They will disagree, and that gap is the entire point of the exercise. Budget for the overlap and the review time; there is no third cost hiding behind them.

Setup session~1 hour

Two systems running3–4 weeks

Reviewing the gap1 meeting

The trial itself€0

Questions, answered.

Do I have to switch off my current tool?+

No. Nothing has to be switched off on day one. Admetrics runs in parallel with your existing tracking for an agreed validation period, and your current setup keeps recording the whole time.

What if my current contract still has months to run?+

That is not a problem. The parallel run needs no cancellation, so you can keep your current tool until its term ends and switch over whenever it suits you. Talk to us about the overlap and we will find a fair arrangement for it.

How long does the parallel run take?+

Typically about four weeks. Setup takes roughly two hours; the rest is collecting enough orders for a fair comparison. Higher-volume stores usually have an answer sooner.

What if the two systems disagree?+

They will, and that is the useful part. The gap is where conversions were going unrecorded or being credited to the wrong channel. Every difference traces back to an individual order, so it stays a conversation about evidence rather than opinion.

Do we start from zero on historical data?+

No. Orders, products, costs and platform spend are backfilled during setup for the periods your systems can provide. Journey-level tracking begins the day it goes live, which is exactly why the parallel run is worth starting early.

Who does the setup work?+

We do. Setup is a guided session of roughly two hours with a success manager. The 21-day trial then runs on your own orders, so your team's only real jobs are agreeing the baseline up front and reviewing the result at the end.