GA4

GA4 and Conversion Tracking for Fintech and Financial Services in London (Canary Wharf)

Fintech and financial services companies in London, and especially around Canary Wharf, spend heavily on Google Ads, Meta and LinkedIn. Yet their tracking is often weaker than a typical Shopify store’s. The reason is the funnel. A visitor reads the marketing site, moves to a separate app or onboarding domain, passes identity checks, and only becomes valuable days later when they fund an account or sign a contract. Standard GA4 setups lose that person at almost every step.

Short answer: fintech tracking in the UK needs four things that most setups miss:

  1. One GA4 property that follows the visitor from the marketing site into the onboarding app (cross-domain or app + web).
  2. Consent handled properly under UK GDPR and PECR, with Google Consent Mode v2.
  3. No personal or financial data leaking into GA4, Google Ads or Meta, through URLs or events.
  4. The real business outcome (verified, funded, approved) sent back to the ad platforms as the conversion, usually through server-side or offline conversion imports.

I’m Jobair Mahmud, a remote GA4, server-side tracking and Google Ads consultant. I’ve set up cross-domain GA4 and consent-aware conversion tracking for a European fintech company, and I work with UK financial services firms and the agencies that run their ads. Below is how I approach it.


Why fintech tracking breaks more often than ecommerce tracking

The journey crosses domains and devices

A typical journey looks like this: brand.co.uk (marketing site) → app.brand.co.uk or a third-party onboarding provider → identity verification (KYC) → mobile app download → first deposit. Every handover is a chance for GA4 to start a new session and credit the conversion to “direct” or to your own domain instead of the Google Ads campaign that paid for the click.

The valuable event happens late and off the website

“Sign-up started” is cheap and plentiful. “Account approved” and “account funded” are rare and valuable, and they usually happen in your back end, CRM or core banking system, not in the browser. If you only send Google Ads the sign-up, Smart Bidding will learn to find people who start sign-ups and never finish them.

The data is sensitive

Onboarding pages contain names, email addresses, dates of birth, income ranges and sometimes account numbers. If any of that ends up in a page URL, page title or event parameter, it lands in GA4 and possibly in Google Ads and Meta. That’s a data protection problem and breaks Google’s and Meta’s own policies.


1. Follow the visitor from the website into onboarding

If onboarding is on a subdomain (app.brand.co.uk), GA4 handles it automatically as long as both use the same measurement ID and the same cookie domain. Check that the subdomain isn’t loading a separate GTM container with a different GA4 setup.

If onboarding is on a different domain (brand-onboarding.com or a provider’s domain), set up cross-domain measurement in the GA4 admin and check the _gl parameter survives every redirect. I explain how in my cross-domain tracking guide and the _gl parameter explainer.

If the journey continues in a mobile app, use one GA4 property with web and app data streams, and pass a user ID once someone logs in, so the web campaign and the app deposit can be connected.

Exclude your own payment and verification providers from referrals, so a visitor returning from an identity check doesn’t show up as a new referral visit.


The rules are the same as for any UK website, but financial firms are more likely to be audited, so they need to be done properly:

  • Analytics and advertising tags must not fire before the visitor consents (PECR). The Data (Use and Access) Act 2025 relaxes this for some analytics-only uses, but advertising tags still need consent.
  • Use a consent platform (OneTrust, Cookiebot, Usercentrics and similar) connected to Google Consent Mode v2, so Google can model conversions from visitors who decline.
  • Make sure the same consent choice carries across the marketing site and the onboarding domain. A visitor who declined on one shouldn’t be tracked on the other.
  • If you use server-side tracking, the server must respect the consent state too. Server-side is not a way around consent.

More detail: Consent Mode v2 setup in GTM.


3. Keep personal and financial data out of analytics and ads

This is where I find the most problems in fintech accounts:

  • Personal data in URLs. Onboarding forms that use GET requests or put emails in query strings (?email=...). GA4 records the full URL. Fix the form, and add GA4 data redaction for email and query parameters as a safety net.
  • Personal data in page titles or events. “Welcome, Sarah” in the title, or a form-field value sent as an event parameter. Review every custom event before it goes live.
  • Financial details sent to ad platforms. Loan amounts, income bands or balances should not be sent to Google Ads or Meta as event data. If you need value-based bidding, send a modelled value (for example an expected lifetime value per product), not the customer’s real figures.
  • Enhanced conversions and Meta CAPI. Hashed email and phone are allowed and useful, but only with consent, only hashed, and only for the conversion event, never on every page.

Server-side GTM helps here because it gives you one place to remove or rewrite fields before anything reaches Google or Meta.


4. Send the real outcome back as the conversion

For a fintech, I usually set up a ladder of conversions:

StageWhere it happensHow it reaches Google Ads / MetaUse for bidding?
Sign-up startedWebsiteGTM tagNo (secondary)
Application submittedOnboarding appGTM or server eventEarly on, until volume builds
Verified / approvedBack end or KYC providerServer-side event or offline importYes
Account funded / first trade / contract signedCore system or CRMOffline conversion import with valueYes, primary

To connect the late stages back to the original ad click, the Google Click ID (gclid), and for Meta the fbc/fbp values, must be captured at sign-up and stored with the customer record. When the account is funded, your system sends that conversion back. I’ve built this for a UK retailer using a scheduled offline conversion import; the same pattern works for fintech, usually with a CRM (HubSpot, Salesforce) or a small server-side job instead of a spreadsheet. See the offline conversion tracking guide.

Google requires consent fields (ad_user_data and ad_personalization) on offline conversions from UK and EEA users, so include them from the start.


5. Financial promotions and ad platform rules

Tracking doesn’t make an ad compliant, but financial firms should know the platform rules that sit around it:

  • Google Ads and Meta both require UK financial services advertisers to complete a verification that checks FCA authorisation before ads for many financial products can run. If your ads are limited, check verification before blaming tracking.
  • FCA financial promotion rules (fair, clear and not misleading; specific rules for cryptoassets and high-risk investments) apply to the ad and landing page content. Your compliance team owns this; my job is to make sure tracking changes don’t alter approved landing pages and that test pages aren’t indexed or advertised.
  • Personalised advertising policies limit targeting around sensitive financial circumstances. Keep remarketing audiences based on product interest, not on someone’s financial situation.

I’m not a lawyer or a compliance adviser. I build tracking to fit the policy your DPO and compliance team set.


6. Server-side tracking hosted in London

Many financial firms prefer data to be processed in the UK. Server-side Google Tag Manager can run on Google Cloud Run in the London region (europe-west2), on your own subdomain such as sgtm.brand.co.uk. That gives you:

  • one controlled place where every event is checked and cleaned before it goes to Google, Meta or LinkedIn
  • first-party cookies that last longer in Safari
  • a clear record of exactly which data each platform receives, which helps when compliance asks

See the UK server-side tracking consultant page and the Stape vs Google Cloud cost guide.


Beyond Canary Wharf: Edinburgh, Belfast and the rest of the UK

The same approach applies to financial services across the UK: asset managers and insurers in Edinburgh, fintech and software firms in Belfast (often selling into the Republic of Ireland in euros too), and insurers in Cardiff. More on each region on my GA4 consultant UK page.


Frequently asked questions

Can GA4 be used by a regulated financial services firm in the UK?

Yes, many do. The important parts are consent before tracking, no personal data in GA4, sensible data retention settings, and a clear record of what is collected. Your DPO decides the policy; the setup should make that policy easy to follow.

Should we send account balances or loan amounts to Google Ads as conversion values?

No. Send a modelled value per product or customer type instead. You still get value-based bidding without sharing a customer’s real financial details.

Our onboarding is run by a third-party provider. Can we still track it?

Usually yes, with cross-domain measurement if the provider allows your tags, or with a server-side event or webhook from the provider when an application is completed or approved.

No. Under PECR, advertising tracking still needs consent whether it runs in the browser or on a server.

Do you work with Canary Wharf and London agencies?

Yes. I work white-label for agencies that run fintech and financial services accounts, under their name and NDA, during UK working hours. See for agencies.


Next step

If your ads are optimising for sign-ups that never fund, or you’re not sure what data your tags send, start with an audit. I’ll review your GTM, GA4 and ad conversions and send a written report with a fixed quote for the fixes.

Direct Implementation White-Label for Agencies

Need this implemented, audited, or white-labeled?

Whether you are an eCommerce brand owner fixing signal loss or a digital marketing agency needing a senior analytics engineer under strict NDA, I deliver 24–48h SLA turnaround with 8.5+ EMQ match rates.

✓ 24–48h Emergency SLA ✓ Strict Non-Solicitation & NDA ✓ 8.5+ Event Match Quality
Jobair Mahmud

Jobair Mahmud

AI-Driven Performance Marketing & Web Analytics Consultant

Helping eCommerce brands and marketing agencies trust their ads data, scale Smart Bidding, and eliminate wasted ad spend through bulletproof server-side tracking and PPC management.