Conversion tracking connects marketing activity to the actions that matter. Attribution adds context about how channels, campaigns, and interactions contributed along the way. Together they move measurement past impressions, clicks, and sessions, toward the question a marketing budget is actually judged on: which activity is producing outcomes worth having.
Define
Agree what genuinely counts as a conversion before arguing about which channel deserves credit for it.
Connect
Carry source, campaign, and landing information as far toward the real outcome as client systems allow.
Interpret
Read performance through more than one attribution lens, and explain where the views disagree.
Qualify
State the limits of the data plainly rather than presenting an incomplete picture as certainty.
Selected clients
A selection of the specialists who lead and shape this work, supported by a broader multidisciplinary team.
Conversion tracking and attribution exist to answer one question: which marketing activities are contributing to meaningful outcomes. Conversion tracking is the recording half. It captures the actions an organization has decided are worth counting, along with the source, medium, campaign, and landing experience that preceded them. Attribution is the interpretive half. It decides how credit for those actions should be distributed across the interactions that came before, which is a harder problem than it sounds, because most conversions follow several.
The shift this service is really making is one of altitude. Impressions, clicks, and sessions describe activity. Source, journey, conversion, and outcome describe contribution. The first sequence can look excellent while the business notices nothing. The second is harder to assemble and considerably harder to argue with, because it ends at something the organization already recognizes as valuable: a lead worth calling back, an application, an enrollment, a booked appointment, a purchase.
This page owns the connection between marketing and outcomes. Deciding what the numbers mean and which questions to ask belongs to marketing analytics. Building the events, tags, data layers, and platform connections that make collection possible belongs to analytics implementation and tracking. Presenting the result consistently is reporting work. Here the concern is narrower and specific: what counts as a conversion, how it is connected back to a source, and how much credit that source can honestly claim.
The channel that closes a conversion is not necessarily the channel that caused it.
We are platform-independent. Tooling is chosen per engagement, and every account is client-owned.
Google Analytics 4Google Tag ManagerGoogle Search Console
Looker StudioSeven failure patterns we see repeatedly, and what we change about each.
If none of these describe the situation, we will say so before a proposal is written.
Media is running across search, social, video, and audio, and the only shared measure is a last click report that credits whatever the person touched last.
Forms and calls both produce inquiries, but nobody can say which campaigns produce inquiries worth having rather than inquiries in general.
The phone is the primary conversion and it is either untracked or tracked in a system that never talks to the advertising platforms.
Platform reported revenue, analytics revenue, and the store's own numbers all disagree, and the gap has never been explained.
Connected TV, programmatic video, geofencing, or out of home is running and there is no agreed way to judge whether it contributed anything.
The outcome is an application, an enrollment, a registration, or an appointment rather than a sale, and the standard ecommerce measurement playbook does not fit.
Most attribution arguments are really definition arguments in disguise. Before anyone can say which channel deserves credit, the organization has to agree on what it is giving credit for, and that agreement is where we start. We work out which actions genuinely represent progress, which are useful context but not conversions, and how far past the website the signal can practically be followed given the systems the client already runs. Then we connect the pieces: campaign and source information carried into the form, the call, the cart, and where possible into the CRM record and back out again as an outcome. From there we read performance through more than one lens, because first interaction, last interaction, and a platform's own model each describe the same journey differently. Where the data cannot support a conclusion, we say so rather than filling the gap with a confident chart.
Attribution is a defensible reading of the signals, not a reconstruction of intent.
Fig. 02
Channels tend to play roles rather than occupy fixed positions in a queue. Connected TV, programmatic video, audio, geofencing, and out of home usually introduce. Display, paid social, and organic search usually inform. Paid search, local search, and direct usually arrive with intent already formed. But the journeys underneath are not uniform: one person is introduced by a video ad and converts through a branded search weeks later, another arrives ready and converts on the first visit, and a third never touches an awareness channel at all. Measurement has to accommodate all three rather than assume the tidy one.
Channel roles are tendencies observed over time, not rules that hold per person.
A channel that rarely closes can still be the reason the journey started.
Seven stages, run in order. Measurement design is agreed before any budget is committed.
Scoped per engagement. Most programs use four or five of the capabilities below.
Working through which actions represent genuine progress toward the objective, which are intent signals worth watching, and which should not be counted at all. The definitions are written down so every platform can be configured against the same list.
Mapping how people actually reach a conversion across paid, organic, local, direct, and referral paths, so the measurement design reflects the journeys that exist rather than the funnel diagram in a deck.
Connecting form submissions to the source, medium, campaign, and landing page that produced them, and carrying that information into the systems where the lead is worked.
Tracked numbers and dynamic number insertion configured so inbound calls can be associated with campaigns and landing pages, with call outcomes reviewed rather than assumed.
Transactions, revenue where appropriate, returning customer behavior, and the relationship between campaigns and purchase, read alongside the store's own reporting rather than instead of it.
Where client systems allow it, carrying acquisition data into the CRM and feeding qualified outcomes back toward the platforms, so lead quality becomes measurable rather than inferred.
A documented source, medium, campaign, and content convention applied across every channel and vendor, so campaigns stop fragmenting into near duplicate labels.
Reading performance through first interaction, last interaction, and platform modelled views, explaining where they disagree and what that disagreement implies for budget.
Fig. 03
One journey, recorded once, read three ways. First interaction credits the connected TV exposure that introduced the brand. Last interaction credits the paid search click that immediately preceded the form. A platform's modelled view distributes credit across the observed touchpoints using its own methodology and whatever signals it can see. Nothing about the journey changed between those three readings. This is why comparing platform totals is usually less useful than understanding which lens each one is applying, and why a single number presented as the truth should be treated with suspicion.
Reading two lenses side by side is often more informative than defending one.
Where the lenses disagree sharply, that gap is itself a finding worth reporting.
Not sure what your numbers are crediting
We will review how conversions are defined and captured today, what acquisition information survives from click to lead, what your platforms are each counting, and where credit is currently being assigned in ways the underlying data does not support.
04
Good marketing begins and ends with reliable data. Clean instrumentation, honest attribution, and reporting an executive team can defend, established before the spend and revisited after it.
HUB International
Professional services. Advertising.
Insurance is one of the most competitive categories in paid search, and HUB International served industries whose needs had almost nothing in common. A business researching employee benefits needed a different message than one evaluating cyber liability, and the budget had to stay clear of claims, careers, and consumer traffic.
Campaign themes structured by insurance need
OctoClean needed to grow its franchise network, but its website was not reaching or educating prospective business owners. Candidates had to understand the opportunity, the support behind it, and what set the model apart from other commercial cleaning franchises.
Franchises | Web
2,247%
Increase in website traffic during the engagement
Franchises
Web
OctoClean
01 / 02
Writing from the team running this work, published as the field changes.
According to Google’s official announcement on March 16, 2022 Google will begin sunsetting Universal Analytics (UA) on July 1, 2023. The
May 26, 2022
How to Make Sure Your Core Web Vitals are Ready for Google’s Algorithm Update
Are The Presidential Polls Wrong?
The Marketing Budget Formula
Why Personalized Advertising is the Future of Digital Marketing
Conversion tracking records the actions that matter, such as leads, calls, appointments or purchases, and connects each one back to the marketing source that preceded it.
Attribution is the framework used to assign credit for a conversion across the marketing interactions that came before it. It is an interpretation of observed signals, not a record of intent.
Conversion tracking records that a meaningful action happened. Attribution decides which marketing interactions get credit for it. Accurate tracking has to come first.
Actions that represent real progress toward the objective, such as leads, calls, applications, bookings or purchases. Lighter interest signals are useful context but should not be counted as primary conversions.
Each platform uses its own attribution model, conversion window, and identity signals, so their totals were never designed to match. Understand what each one counts and choose one reference source per question.
First interaction credits the touchpoint that introduced someone. Last interaction credits the one immediately before converting. Both are partial views, and reading both is usually more informative than choosing one.
No. Attribution reconstructs journeys from incomplete signals. It can show consistent patterns of which sources introduce, assist, and close conversions, which is enough to guide budget decisions, but not causal certainty.
Tracked phone numbers, often swapped dynamically based on how a visitor arrived, let an inbound call be connected to its source, campaign, and landing page. It has limits and cannot capture every call.
Often partially, when client systems allow it. Acquisition data is carried into the lead record and outcomes are fed back where an import path exists, which lets lead quality be judged rather than just lead volume.
Not by last click alone, and not by assumption. Use assisted conversion patterns, movement in branded and direct demand, and market level comparisons designed before the campaign runs.
UTMs are link parameters that tell analytics where a visit came from. Inconsistent naming splits one campaign across several labels and fragments the data, so a documented convention is part of disciplined measurement.
Conversion tracking and attribution
Tell us what you are running, what you count as a conversion today, and where the numbers stop making sense. We will look at how conversions are captured, what acquisition data survives the journey, and what could reasonably be connected given the systems you already have.
Call (800) 505-7570.
