Paid advertising · Analytics

Supermetrics integration — full-funnel
visibility for paid advertising

A fintech paid team was optimising for sign-ups. The business cared about funded accounts. We connected the dots — joining ad spend data with the full client funnel in BigQuery, then piping it back into their advertising stack via Supermetrics.

0
Manual data exports — fully eliminated from the workflow
3
Funnel stages connected — application, funding, trading
Full funnel
Ad click to funded account — visible for the first time
Industry Fintech · Paid advertising & growth
Engagement type Data modelling · Advertising analytics
Stack

The paid team was flying blind past the sign-up

The paid advertising team at a growth-stage fintech knew which campaigns were driving sign-ups. They had no idea what happened after. The data stopped at the top of the funnel — click, impression, conversion — and there was no way to trace a user from an ad through to application, funding, or live trading activity.

As a result, the team was optimising for cost-per-click and cost-per-sign-up. The business cared about cost-per-funded-account. Those metrics told completely different stories about which campaigns were working — and without the downstream data, there was no way to know which channels were actually driving revenue.

A campaign could be delivering cheap sign-ups that never converted to funded accounts — and the paid team would have no signal to act on. Budget was being allocated on incomplete data.

Full-funnel data layer, connected to the paid stack

We modelled the complete client funnel in BigQuery — from application through funding to trading activity — and joined it with client attributes including the demographic and contact data needed for meaningful audience segmentation. Then we connected it all to the paid advertising stack via Supermetrics.

01
Full-funnel data modelling

Built dbt models in BigQuery covering every stage of the client lifecycle — application status, funding events, and trading activity — as a unified analytical layer. Each stage is a clearly defined model with timestamps and statuses, giving the paid team a clean, queryable view of where every acquired user ended up.

02
Client attribute enrichment

Joined the funnel models with client-level attributes — demographic data, contact details, product preferences, and lifecycle signals. This enriched layer gave the paid team the audience depth needed for precise segmentation, lookalike modelling, and exclusion lists based on real product behaviour.

03
Supermetrics BigQuery connector

Configured Supermetrics to pull from the enriched BigQuery layer directly into the paid advertising platforms. The connector eliminated manual exports entirely — data flows automatically, the paid team works from live funnel data, and any new campaign analysis starts from the same trusted model the analytics team uses.

Optimising for the metric that actually mattered.

Full funnel
Paid team gained end-to-end visibility — from ad click to funded account — for the first time
0
Manual exports — Supermetrics connector replaced the entire manual workflow
Revenue
Optimisation target shifted from sign-ups to first funded account — aligned to actual revenue
Paid team gained full-funnel visibility — from ad click to funded account — for the first time, without any manual data work
Optimisation target shifted from top-of-funnel sign-ups to first funded account, directly aligning ad spend with the metric the business cared about
Richer audience data enabled more precise targeting, lookalike segmentation, and exclusion lists based on real product activity — reducing wasted spend
Supermetrics BigQuery connector eliminated manual exports from the workflow entirely — live funnel data flows directly into the advertising platforms on schedule

Model the funnel first. Connect the tools second.

The Supermetrics connector was the easy part. The hard part — and the part that made everything else possible — was modelling the full client funnel in a way the paid team could actually use. That meant working with the product and analytics teams to agree on what "funded account" meant in the data, how to handle partial onboarding, and which funnel stages mapped to which campaign goals.

Once the funnel model was trusted, connecting it to the paid stack was straightforward. The paid team got live data. The analytics team kept their source-of-truth model. And both teams were working from the same definitions — which meant no more conflicting numbers in campaign reviews.

Ready when you are.

Book a free 30-minute call. We'll find the gaps, prioritise the fixes, and give you a clear path forward.

Usually responds within 24 hours