Primal, a marketing agency built for the AI era, replaced hours of hopping between ad managers with one blend in Supermetrics. Reporting time for the operations team dropped from roughly two hours to about 30 minutes, a reduction of roughly 75%. Wisantoon J. (Sun), Data Analytics Lead, set up the first data blend himself in about 30 minutes, and the operations team now duplicates that master blend and swaps in each client's accounts.
Their challenge
Primal is a digital marketing agency built for the AI era, with teams in Bangkok, Kuala Lumpur and Singapore. It is a Google Premier Partner and a Meta Business Partner. Primal brings strategy, search, performance marketing, creative, social and data together in one integrated team. It works with more than 300 brands across 40+ industries, and every one of them has its own budget, its own pacing and its own reporting rhythm.
"Four platforms for one brand is a task," says Wisantoon J. (Sun), Data Analytics Lead at Primal. "Four platforms for hundreds is a data problem." Sun solved it by defining each blend once in Supermetrics and reusing it across every account.
Clients ask Primal to work across whatever ad platforms they need, set a conversion goal and a budget, and then optimize toward the KPIs. Tracking whether each campaign was on pace meant a routine that repeated for every media client, every week:
- Open four or more ad managers (Meta Ads, Google Ads, TikTok Ads and Microsoft Ads), one at a time, for one client at a time.
- Collect the numbers for the period and retype them by hand into a Google Sheet.
- Work out pacing manually, comparing spend to date against the monthly budget.
- Act on a sheet that was already a day old when it was finished.
Each Primal ad specialist looks after several brands, so every manual step was repeated many times over.
The second problem was vocabulary. Every platform names the same metric differently. Spend is "amount spent" on Meta Ads, "cost" on Google Ads and TikTok Ads, and "spend" on Microsoft Ads. Someone had to match them by hand for every report, and similar names left room for human error. "The names are different, but it's all the same thing: spend," says Sun. "That's the real blocker."
The workarounds that didn't hold
Primal looked at several routes before landing on blending.
- A dashboard per platform. Four dashboards still don't give a cross-platform total or a pacing view.
- Building it themselves. A schema map for each platform, maintained forever and rewritten every time an API changes.
- Blending inside Looker. The numbers didn't always reconcile with the platforms.
- Master Google Sheets. Sun pulled data in with the Supermetrics Google Sheets destination and tied it together with formulas, with a template each team member configured from the sidebar. It was a step forward, but it was still manual: formulas broke when someone edited a cell, so Sun had to lock the master sheets, and every client with a different goal, such as conversions rather than top of funnel, needed its own rework.
The result was accurate on the day each sheet was built, and fragile after that.
What they did
Sun moved the harmonizing out of spreadsheets and into Supermetrics data blending. He built one union blend across every ad platform a client runs, mapped each platform's fields to a single canonical name once, with no code, and pointed everything downstream at the result.
Supermetrics sits in the middle of Primal's setup. On one side are the channels Primal works across: performance, search (SEO and AI search), CRO and social. On the other are the destinations the team reads from, Google Sheets and Looker.
Defining every blend once
In the blend setup, Sun picked the dimensions and metrics he wanted to align and gave each a global name that every platform, department and client now shares. The dimension blend harmonizes 56 dimensions into one set of names. The metric blend has one row per metric and one column per source, so the four different names for spend all map to a single "spend" that everything downstream reads.
Setting it up took Sun about 30 minutes. "I set it up by myself, learning as I went," he says. "It took around 30 minutes."
One blend, one dashboard
The blend lands in BigQuery, and Looker reads that single table. Every chart, for every account, means the same thing, and spend, revenue and ROAS follow one definition whatever mix of platforms a client runs. Clients can open their dashboard any day they like, and how often they look depends on the client.
Handing it to the operations team
Once Sun had checked that everything was collected correctly, he handed the master blend to the operations team. They duplicate it, swap in the client's accounts, and check that the right accounts are selected. Every client ends up on the same standard, and nobody has to rebuild a blend account by account.
It is also safer. With no formulas in a shared sheet for someone to overwrite, the logic sits in the back end. "If we blend from the very beginning, at the source, it's much safer for the team to use over the long term," says Sun.
Outcome and impact
Primal's operations team now starts from one blend instead of four ad managers.
- 2 hours down to 30 minutes: the time spent gathering data across platform interfaces fell by roughly 75%, and now goes on a quick check of the duplicated master blend.
- One setup, reused everywhere: the first blend took about 30 minutes to build, and the same mapping is reused for every account instead of being rebuilt one at a time.
- 384 accounts on one definition: spend, revenue and ROAS mean the same thing across the four ad platforms Primal's clients run today, with 56 dimensions harmonized into one set of names.
- Around 300,000 rows harmonized automatically: a year of daily platform data, September 2025 to September 2026, unified without anyone renaming a field by hand.
- No code, no broken formulas: the logic lives in the back end, so a team member editing a cell can no longer break the numbers.
- Benchmarking by sector: because 300+ brands across 40+ industries now sit on one definition, Primal compares each client with the benchmarks for its own sector and optimizes against them.
The saved time goes to higher-value work. "Our operations team is more efficient," says Sun. "They get that time back to spend on things that are more valuable, like ad strategy and optimizing better, instead of setting up and tracking everything manually."
For Sun, the bigger gain is one agreed number. "It isn't the hours saved," he says. "It's that the argument about whose number is right ends."
What's next
With reporting unified, Sun's next question is control. Every Primal client has its own goals and spend limits, and he wants Supermetrics to flag when a campaign passes one, for example when Google Ads spend climbs past an agreed ceiling, without setting that up outside the platform.
That is what Supermetrics Business Context and monitoring are for. They hold each client's thresholds, goals and KPIs, and alert the team by email or Slack when spend gets close to a limit or a campaign underperforms. Supermetrics Campaign Management then lets the team pause a campaign, add budget or launch one from an AI chat, with an approval step before any change goes live.
Primal already uses the Supermetrics MCP server with Claude. Sun plans to test these features with a small group of clients.
One number from every ad platform
Blend your ad data once in Supermetrics, map every platform's fields to a single name, and point every account, dashboard and report at the same table. No code to maintain.