Why this case study is different
Style Strand Media is our own agency.
We are telling this one in the first person because pretending otherwise would be dishonest, and because the fact itself is the point: every method ANtreliX sells was run on our own business before a client paid for it. ANtreliX exists because this engagement worked.
Challenge
The agency had the problem most agencies have, and it is not a marketing problem.
Value was measured in deliverables — articles published, campaigns launched, posts scheduled — because that was what could be counted. Meanwhile each client's data lived in a different place: ad platforms, GA4, Search Console, the email platform, a CRM or property management system, call logs, social platforms. Nothing talked to anything else.
Reporting was therefore a manual assembly job. Someone exported, pasted, formatted and wrote a monthly PDF that was already out of date the day it arrived — and it could describe channels one at a time, but never a campaign end to end. When a client asked "did that work?", the honest answer took days to assemble and often arrived after the decision had been made.
Underneath that, production was eating the margin. Research, briefs, drafts, SEO elements, publishing checklists, image handling — all of it done by hand, all of it repeatable, none of it automated. The strategic conversations that actually justify an agency's fee were being crowded out by throughput.
Approach
Three moves, in a deliberate order: join the data, make reporting live, then automate the repeatable work.
The sequence matters. Automation applied before measurement just produces waste faster, and nobody can tell. Measurement first meant every automation that followed could be judged.
Breaking the data silos
- An inventory, per client, of every data source and what question each one can actually answer.
- A standard campaign and UTM taxonomy applied across clients, so channels can be compared at all — the unglamorous step most agencies skip, and the reason most agency reporting cannot be trusted.
- Scheduled pulls into one store per client: platform connectors where they exist, scripted exports where they don't.
- Locked definitions. What counts as a lead, a booking, an assisted conversion — one definition, used the same way for every client, every month.
- Attribution rules that state plainly what the data can and cannot prove, so a report never claims credit it hasn't earned.
Live reporting instead of monthly PDFs
Looker Studio (formerly Google Data Studio) became the reporting layer, with a standard dashboard template per client type — vacation rentals, real estate, home services, professional practices — so a new client is set up in hours rather than reinvented from scratch.
- Clients get a link, not an attachment. The data refreshes on its own; there is no "reporting week".
- Every dashboard answers the same five questions: what happened, compared to when, what drove it, what is coming, and what needs a decision.
- A short written narrative sits on top of the live data each month. That is the part a dashboard cannot do, and it is where the agency's judgment shows.
- Internal dashboards track the agency itself: publishing cadence, work in progress, and performance across the whole client portfolio.
The second-order effect was the valuable one. When a client can see performance any day, the monthly meeting stops being a recap and becomes a decision.
Automating the repeatable work
- Standard operating procedures for AI-assisted research, drafting, SEO element sets, image handling and publishing — written down, so quality does not depend on who is working.
- Production pipelines per content type: video transcript to multi-channel asset suite, full article packages with a complete SEO element set, Business Profile posts, newsletter sections.
- Automated assembly and scheduling of reporting, publishing checklists, and validation scripts that check headings, structured data and links before anything goes live.
- Human review as a gate that is never skipped. AI multiplies output; it does not get the last word.
The portfolio redesign
With measurement and production under control, the service list could change from what the agency produced to what clients actually buy: business intelligence and dashboards, predictive analytics, AI adoption roadmaps, workflow automation, and AI search visibility — alongside the content and SEO work that funds it.
That redesign is what eventually separated into ANtreliX.
Results
Reporting became interpretation, not assembly. The hours that went into building reports now go into reading them and deciding what to do.
Clients see their performance continuously. No waiting for the tenth of the month, and no argument about whose numbers are right — there is one source.
Campaigns can be judged end to end. With sources joined and definitions locked, a campaign can be followed from impression through to booking or enquiry rather than described channel by channel.
Margin moved from production to strategy. Automating the repeatable work freed the time that strategic engagements require — which is also the work clients value most.
The method became a product. Every client system built since descends from this: the vacation rental intelligence platform, the brokerage lead reporting, the AI visibility audit. ANtreliX is the productized version of what we built for ourselves.
The analytical discipline came from here. The definitions, the attribution honesty rules, and the refusal to report correlation as causation were established in our own reporting first, and now govern every client analysis we produce.
Technologies and methodology
- Looker Studio dashboards with per-industry templates
- GA4, Google Search Console, ad platforms, email platforms, CRM and property management system exports
- Python and Pandas pipelines with scheduled refresh; SQL storage
- Standardized campaign and UTM taxonomy; locked metric definitions
- AI-assisted production workflows with documented SOPs and human review
- Automated QA validation for structure, schema and links
- Portfolio-level internal reporting
Why it worked
Because the order was right. Most agencies buy a dashboard tool first, point it at data nobody has cleaned, and end up with a prettier version of the same confusion. Joining the sources and agreeing what the words mean is unglamorous work that pays for everything built on top of it.
And because it was done on our own business, where the cost of getting it wrong was ours to carry.
Key takeaway
An agency that cannot measure its own work has no business promising to measure yours. We rebuilt ours first — and that is the only reason we are willing to sell the method.