EAGLEXOR

Analytics & tracking

Measurement your team can trust and act on.

We agree which customer actions matter, configure analytics and tags to record them, check that the data is right, and set up reporting your team can use to make decisions.

See what we configure
  1. Define what matters
  2. Configure measurement
  3. Validate data
  4. Interpret results
Two colleagues at a table at night agree on a row of a blank sheet, one pointing while the other holds a pen

Illustrative image

Good reporting starts with definitions everyone agrees on.

Questions worth investigating.

Most measurement problems are quiet. Reports keep loading, but the numbers drift away from what the business sees. These are the issues we look for first. Not every setup has them, and the audit shows which ones yours does.

  1. Are conversions being recorded at all?

    What you may notice

    Orders, leads, or sign-ups in your store or CRM have no matching event in analytics or the ad platforms.

    What we check

    Whether the event fires, reaches each platform, and is marked as a conversion where it should be.

  2. Is the same action counted twice?

    What you may notice

    Analytics shows more purchases or leads than the orders and enquiries you actually received.

    What we check

    Duplicate tags, reloaded confirmation pages, and browser and server events sent without deduplication.

  3. Does everyone mean the same thing by a conversion?

    What you may notice

    Teams quote different numbers for the same week, each taken from a different report or platform.

    What we check

    Event and conversion definitions, naming, and which platform each figure in a report comes from.

  4. Can you tell which campaigns brought customers in?

    What you may notice

    Visits arrive as direct or unassigned, or campaign names change from one team or launch to the next.

    What we check

    UTM conventions, ad platform auto-tagging, redirects that drop parameters, and channel grouping.

  5. Do the reports answer the questions you ask?

    What you may notice

    Dashboards exist, but decisions still rely on screenshots, exports, and educated guesses.

    What we check

    Which business questions matter, and whether the data needed to answer them is collected at all.

What we set up, step by step.

The work follows the same four steps every time. Choose a step to see what it covers and what you receive at the end of it.

Hands line up a short column of blank cards along a violet line and push a loose pile aside

Illustrative image

Step 01

Define what matters

Before any tag changes, we agree what the business needs to measure and what each event means.

  • Measurement planning

    Your goals turned into a short list of customer actions worth measuring, and the questions each one should answer.

    You receiveMeasurement plan

  • Event and conversion definitions

    A name, trigger, and parameters for every event, and a decision on which events count as conversions in each platform.

    You receiveEvent specification

  • Campaign tagging conventions

    UTM naming rules and ad platform auto-tagging, so campaigns arrive in reports under consistent names.

    You receiveUTM naming guide

Measure the journey, not just the sale.

A purchase is the last of several steps. Measuring the steps before it shows where people stop, though each one answers only part of the question. These are the stages we usually measure in an online store.

Plan your journey events
  1. A shopper leans in to look at a pair of black headphones on a lit display stand

    Product viewed

    • view_item_list
    • select_item
    • view_item
    What it can show
    Which products and lists get attention, and which campaigns brought those visitors.
    What it cannot show alone
    Whether the visitor liked the product, or why they moved on.
  2. A hand places a boxed pair of headphones into a wire shopping basket

    Added to cart

    • add_to_cart
    • remove_from_cart
    • view_cart
    What it can show
    Which products people intend to buy, and how often carts are left behind.
    What it cannot show alone
    Why a cart was abandoned. That takes research, not just events.
  3. A woman holds a payment card while typing on a laptop, headphones resting beside it

    Checkout

    • begin_checkout
    • add_shipping_info
    • add_payment_info
    What it can show
    How far people get through checkout, and which step loses them.
    What it cannot show alone
    Steps inside a hosted checkout that your platform does not let tags see.
  4. Hands lift the lid of a delivery box to reveal a pair of headphones in tissue paper

    Purchase

    • purchase
    • refund
    What it can show
    Revenue, order value, and items, linked to campaigns where the data allows, and refunds where they are sent.
    What it cannot show alone
    Every customer's full path across devices, or sales made offline.

Not an online store? The same approach covers form submissions, bookings, calls, and sign-ups, measured with the steps that lead up to them.

The event names shown are Google Analytics 4 recommended events. Your plan may name events differently where a platform requires it, and only includes the ones your reports need.

From audit to handover.

Each step ends with something written down, so you can see what changed and why. If parts of your setup already work, we keep them.

Start with a tracking audit

  1. Audit

    We review your current analytics, tags, conversions, and reports against how your business works, using read access wherever possible.

    OutputAudit findings, ranked by impact

  2. Measurement plan

    We agree the events, conversions, parameters, and naming your reporting needs, and write them down before anything is built.

    OutputMeasurement plan and event specification

  3. Configuration

    Tags, events, and conversions are built or corrected in your analytics, tag manager, store, and ad platforms, with every change recorded.

    OutputConfigured tags, events, and conversions

  4. Validation

    Each event is tested for firing, values, and parameters, and compared with store or CRM records. Problems are fixed before handover.

    OutputValidation log and fixes

  5. Handover

    Your team receives the documentation, access, and reporting views, with a walkthrough of how to use and maintain them.

    OutputDocumentation and reporting handover

Check the data before anyone relies on it.

A report is only as reliable as the events behind it. These are the checks we run before any reporting is handed over.

Check your data quality
A magnifying glass held over two overlapping copies of the same receipt beside coins in different currencies

  1. Missing events

    What we look for
    Key actions with no matching event, or events that stop firing on some pages, browsers, or devices.
    How we confirm it
    Walking through each journey in debug tools and comparing event counts with store or CRM records.

  2. Duplicate firing

    What we look for
    The same purchase or lead recorded twice by repeated tags, reloaded confirmation pages, or browser and server events without deduplication.
    How we confirm it
    Checking transaction and event IDs, and reloading confirmation pages to see what fires again.

  3. Incorrect values

    What we look for
    Revenue that treats tax and shipping inconsistently, wrong quantities, or prices sent in the wrong format.
    How we confirm it
    Comparing event values with the matching orders, using test orders only where you agree to them.

  4. Currency handling

    What we look for
    Orders in several currencies reported as one, or values sent without a currency code.
    How we confirm it
    Checking that every value carries the right currency code, and how each platform converts it.

  5. Inconsistent parameters

    What we look for
    The same product, campaign, or form described with different names or IDs across events and platforms.
    How we confirm it
    Comparing parameters with the event specification and correcting them at the source.

  6. Consent behaviour

    What we look for
    Tags that fire before a visitor has made a choice, or that ignore the choice once it is made.
    How we confirm it
    Testing each option your consent banner offers and recording what every tag does in response.

Platforms will still report different numbers. Each one counts, attributes, and times conversions in its own way. Validation explains the differences you should expect, rather than forcing the figures to match.

What you receive at handover.

Everything is written down and handed to your team, so the setup keeps working and can be maintained by whoever looks after it next.

Agree your reporting

  1. Measurement plan and event documentation

    What each event means, when it fires, and which parameters it sends.

  2. Configuration notes

    What was set up or changed in each platform, and why.

  3. Validation findings

    What was tested, what was fixed, and anything still open.

  4. Reporting views

    Reports or dashboards built around the questions you agreed, with the source of every figure labelled.

  5. Recommendations

    What to fix, measure, or test next, in order of impact.

A slim binder with a key on top is handed across a table into another person's open hands

Illustrative image

The setup should outlast the project that built it.

Attribution

The limits we state plainly.

No setup measures every visitor or every sale. Consent choices, browser privacy features, ad blockers, people switching devices, and each platform's own attribution rules all leave gaps. We document where those gaps are and help your team read reports with them in mind.

  • Attribution models estimate how much credit each channel gets. They do not prove what caused a sale.
  • Analytics, ad platforms, and your store will report different totals, and we explain why.
  • Server-side tracking is used only where it is supported and justified. It does not remove these limits.
  • A visitor's consent choice is respected, never worked around.

Analytics questions, answered plainly.

Useful to have ready

  1. Access to your analytics, tag manager, and ad accounts, or the person who manages them
  2. The store or website platform you use
  3. The reports your team relies on today
  4. The decisions you want the data to support
Ask about your setup

  • Usually access to your analytics property, tag manager container, ad accounts, and store or website platform. We start the audit with read access where possible and ask for edit access only where changes are agreed. Access can be removed when the work ends.

  • No. The audit shows what is working and what is not. We keep what is sound, fix what is broken, and rebuild only where the existing setup cannot be repaired.

  • Yes. On Shopify we use its customer events where relevant and work within what its checkout lets tags see. For other platforms, we confirm what can be tracked during the audit, because each one limits tracking differently.

  • Yes. We configure conversions in each platform, deduplicate browser and server events where both are used, and check that each platform receives the values it needs. Each platform still attributes conversions by its own rules, so their totals will not match exactly.

  • Tags are set up to follow the choices your consent banner records, with Google consent mode where it applies, and we test that behaviour. Your consent policy, banner, and legal obligations remain your decision, so legal questions are best taken to your own adviser.

  • Not always. It can make some conversion data more reliable where platforms support it, but it adds cost and maintenance, and it does not make attribution complete. We recommend it only where the audit shows it is justified.

  • Ongoing maintenance and regular reviews can be agreed if you want them. Either way, the documentation lets your own team or another provider maintain the setup.

  • It depends on the number of platforms involved, the state of the existing setup, and how quickly access is granted. We give you a timeline after the audit rather than before it.

Start your Analytics & Tracking enquiry

Tell us what needs to be done.

Share the bottleneck, the timing, and what a useful outcome looks like. We review the context and reply with a useful next step.

Prefer email? hello@eaglexor.com

Required fields are validated before submission.