Skip to content

Product

We do the part that vendors leave to you.

Buying analytics software is easy and it is not the hard part. The hard part is deciding what to measure, wiring it so the data is correct, and shaping the reporting so someone actually uses it. That is the engagement.

The sequence

Four steps, in this order.

  1. 1

    Scoping call

    Thirty minutes on your site, your traffic, and the one conversion that matters.

  2. 2

    Written scope

    What we will deploy, what we need from you, the timeline, and what it costs.

  3. 3

    Deployment

    Collection installed on your setup and verified against live traffic before anything is reported.

  4. 4

    Configuration

    Event definition, scoring, source ranking, and reporting built around your team.

01

Why implementation is where it fails

Analytics vendors sell software. The implementation is left to the buyer, on the reasonable-sounding basis that only the buyer knows their business. In practice the implementation is then done by whoever has capacity, against a deadline, using defaults — and the defaults measure traffic, because traffic is what a general-purpose tool measures when nobody has told it what the business produces.

Nothing visibly breaks. The dashboard populates, the numbers move, and the reporting is quietly answering a different question from the one anyone is asking. It usually takes a year and a budget argument for someone to notice, and by then there is a year of data measured the wrong way.

The fix is not more software. It is a few days of skilled configuration work at the start, done by someone who has seen how this goes wrong before. That work is what we sell.

02

What we need from you

Vagueness here loses deals and wastes everyone's time, so here it is specifically.

  • One technical contact who can make changes to how the site is served, available for roughly an hour during deployment and reachable for questions during verification.
  • Enough access to install collection on your setup and verify it against live traffic. The specifics are in the written scope before you commit to anything.
  • A decision-maker who can settle what the conversion event is. This is the one that actually blocks projects — not access, not engineering time.
  • Where the conversion completes inside a third-party flow, someone who knows how that integration was built.

We do not need access to your CRM, your customer database or your payment provider. If a conversion is only confirmable in one of those systems, we will tell you what that means for the measurement rather than asking for the keys.

03

What we do

Deployment is installation and verification. Collection goes onto your setup and then we watch it against real traffic until we are satisfied it is correct — which is a step that gets skipped constantly and is most of the difference between data you can trust and data you assume.

Verification means checking the things that quietly go wrong:

  • That every path to the conversion actually fires it, including the ones through third-party checkout and booking flows.
  • That terminal events survive the navigation that immediately follows them.
  • That sessions and visitors are being grouped the way you think, across the entry points your traffic actually uses.
  • That the numbers reconcile against a source you already trust, so the first question anyone asks has an answer.

Configuration is the second half: defining the conversion event precisely enough to count, wiring the scoring against it, ranking your sources on that axis, and building the reporting around how your team works. That is covered in more detail on the reporting page.

04

Timeline

Expressed as ranges, because a fixed promise before seeing your site would be a guess dressed as a commitment.

Installation and verification is usually days. Configuration is anywhere from a few days to a few weeks, and the variance is almost never technical. It is how settled the conversion definition already is inside your business. Where marketing and finance already agree what counts as a qualified lead, this moves quickly. Where defining it surfaces that they never agreed, the definition work becomes the project.

You get a specific range in the written scope after the call, against your actual site rather than a generic timeline.

05

After go-live

Reporting keeps running and gets refined. The conversion event changes when the business changes. New campaigns need their entry points wired in. A quarter into using it, most teams discover that the ordering they thought they wanted is not the ordering they actually read, and that gets adjusted.

This is the part that separates a deployed system from a delivered project. A configuration that was correct at launch and never touched again describes a business that no longer exists.

06

Why we cap concurrent deployments

The people who take the scoping call are the people who deploy, configure and review. There is no implementation team behind them and no junior tier doing the configuration.

That means a hard limit on how many deployments can be in flight at once. It is a real constraint on the business and we would rather state it than manufacture urgency out of it. Practically, it means there is sometimes a wait, and it means we say no to work that is not a fit — usually on the first call.

Questions

Common questions.

How much of our engineering time does this take?

One technical contact and roughly an hour of their time for the deployment itself, plus availability for questions during verification. The bulk of the work is ours. Where it gets longer is when conversion events complete inside third-party flows — a hosted checkout, an embedded booking widget — because instrumenting those needs someone who knows how that integration was built.

How long does a deployment take end to end?

Installation and verification is typically days rather than weeks. Configuration is the variable part and depends almost entirely on how clearly your conversion event is already defined in the business. When it is well understood, the whole thing is short. When defining it surfaces a disagreement between teams about what counts, that conversation is the project, and it is worth having.

Do you need access to our production systems?

We need enough access to install collection on your setup and to verify it against live traffic. What that means specifically depends on how your site is hosted and built, and it is set out in the written scope before anything starts. We do not need access to your customer database, your CRM, or your payment provider.

What if our site is being rebuilt soon?

Tell us on the call. Deploying measurement immediately before a replatform usually means doing it twice, and the second time is not free. Depending on the timeline the honest answer is sometimes to wait, and sometimes to deploy now specifically so you have a baseline to compare the new site against.

Why do you cap how many deployments run at once?

Because the same people do the deployment, the configuration and the ongoing review. There is no implementation team to hand off to and no junior tier doing the configuration work. Taking on more than we can configure properly would produce exactly the outcome this product exists to prevent.

What happens if we want to stop?

The engagement ends on the terms in the agreed scope. Collection runs on your own setup, so ending an engagement is not a hostage situation — the specifics of what happens to collected data are covered contractually before you start.

Find out what deploying this would involve for you.

Thirty minutes, and a written scope afterwards if it looks like a fit.