Revenue attribution

Add up what your ad platforms claim. It's more than you sold.

Google takes credit. Meta takes credit for the same customer. Your accounting system has the only number that's actually true, and nothing you own connects the two.

Every marketing lead I know is guessing. I was too.

I'm Ross. I run sales and marketing for a Colorado manufacturer that sells six-figure vehicles. Our sales cycle runs months, sometimes most of a year.

Every attribution tool I tried lost the thread somewhere between the first ad click and the deposit. Not because they were badly built, but because they were all watching the browser, and the browser forgets. So I built one that watches the money instead.

This is that system. I run my own budget on it.

None of this is new. We gave it away.

I built my first website in 1995 and I was buying internet advertising before Google Ads existed. Back then the only analytics you had were your own server logs. Crude, ugly, and completely yours. Every request, on your machine, no middleman.

Then we spent twenty-five years trading that away. Third-party cookies, ad networks, platform pixels, tag managers. All of it easier, none of it ours. We handed the record of our own customers to the companies selling us the ads and called it progress.

Now Safari, iOS and ad blockers are quietly repossessing what's left, and everybody is discovering they never owned any of it.

So this isn't a new idea. It's the old one, rebuilt properly. First-party data on your own domain, joined to your own revenue, running on infrastructure that went live last year.

Your ads are optimized for the wrong thing.

You told Google a conversion is a form submission, so Google got very good at finding people who submit forms. Some of them buy. Most of them were never going to.

The platform cannot tell the difference, because nobody ever told it which forms turned into money. So it keeps buying you more of both, and the longer your sales cycle runs, the worse the error compounds.

That isn't a reporting problem you can fix with a better dashboard. You are paying Google every day to go find you the wrong customers.

What this looks like in practice

One deal, tracked the whole way.

1. March. Someone clicks a Google ad. Reads two pages. Leaves. No form, no name, nothing a normal analytics setup will ever connect to anything.
2. May. They come back directly, having thought about it for eight weeks. Most attribution tools now call this a brand new visitor from direct traffic.
3. June. They configure a build and book a consultation. Now there's a name and an email, and the March ad click is still attached to it.
4. August. Deposit received. Six figures, in the accounting system, with a date.
5. That night. The real deal value goes back to Google, matched to the original click. The bidding algorithm learns what an actual buyer looks like.

Every step between March and June is invisible to a browser pixel. That's the part that was costing you money.

Results

I'll show you the actual numbers on the call.

I know precisely how far the platforms overstated their own performance against closed deals, what share of revenue traced back to a first touch more than three months earlier, and how much of the budget moved once we could see it.

Those are real numbers from accounts I run, and I'll walk you through them properly. I'm not going to publish my employer's performance data on a marketing page, and you wouldn't want to work with someone who would.

How it works

Your CRM already knows what closed. Your accounting system already knows what got paid. Nothing here invents a number. It connects the ones you already trust to the ads that produced them.

Beyond the URL

The platforms learn who to find. Your content shows you what to say.

Most attribution stops at the page address. Ours sees inside the page: which sections people actually read, where they slowed down, what they scrolled past, how long they spent in the configurator.

So for every deal that closes, we can replay the path that produced it and score each page by how often it genuinely assisted, and where it sat in the sequence. That separates the pages that move people toward buying from the pages that merely get traffic.

Then it goes one level further. Section-level engagement shows which specific blocks converting visitors consumed. The comparison table they kept returning to. The pricing section they read twice. The testimonials they scrolled straight past.

Two things fall out of that, and both are spendable. You find out which landing pages actually deserve budget, and you find out which arguments earn the sale so you can lead with them everywhere else.

Most companies discover their highest-traffic page contributes almost nothing to revenue. That's usually the first thing we can tell you.

Two things people always ask

Privacy

Count how many companies are watching your customers right now.

Your site probably loads somewhere between fifteen and forty third-party scripts. Analytics, ad pixels, chat widgets, heatmaps, session recorders, a tag manager quietly loading more tag managers. Every one of them sees your visitor. Most join that behaviour to a profile assembled across thousands of other websites. Some sell it onward.

You have contracts with maybe half of them. You cannot say what any of them actually collect. And if a customer asks you to delete their data, you cannot honestly promise that it's gone.

That's a liability sitting on your site today, and nobody is counting it.

We shorten the list. One tracker, on your own domain, writing into storage you control. Conversions go to the two or three ad platforms you already buy from, as confirmation of things that already happened. No data brokers. No third-party enrichment. No profiles assembled across other people's websites.

To be precise about it, because vague privacy claims are their own red flag: the identifiers we send back to Google and Meta are hashed, not anonymous. That's the whole point of a conversion upload. What changes is how many companies hold a copy of your customer's behaviour, and what they're permitted to do with it.

Matched on hard keys. Never on guesswork.

Identities merge on one thing: the normalized email or phone number a person gave you themselves.

No device fingerprinting, no behavioural matching, no statistical inference that two visitors are probably the same human. That approach welds strangers together, and a customer record that quietly contains two people is worse than no record at all.

It also means everything here rests on data your customers handed you directly, which is a much shorter conversation with your legal team.

Opt-outs that actually reach the ad platforms.

A consent banner stops scripts loading in the browser. It does nothing about data already sitting inside a dozen vendor systems, and it has no way to reach back into Google or Meta.

• Global Privacy Control read at the tracker, so the signal is honored before anything gets recorded rather than after.
• Do Not Sell or Share enforced in the pipeline, so an opted-out customer's conversions never upload to any platform.
• Deletion carried out against your store and confirmed, so you can answer the request truthfully.
• A data processing agreement, so we are a service provider under contract, not one more company holding a copy of your customers.

Your customers end up safer and your exposure gets smaller. Those two things are normally a trade. Here they aren't.

Who this is not for

If your customers buy in one session, you don't need this and I'll tell you so on the call. Browser attribution works fine over short windows.

This is built for long sales cycles, high ticket values, real paid media budgets, and a CRM that actually gets used. The longer it takes your customers to decide, the more it's worth.