Every format that matters
PDFs, plain text, manual notes, emails, and scraped URLs with weekly diffing, so a quiet landing-page rewrite never slips past unnoticed.
Context Hub
If ROAS drops, you’ll know why.
You know the call. ROAS is down, and the honest answer takes 45 minutes of clicking through Google Ads, GA4, Slack and the client’s Notion. Give ROASt your briefs, promos, landing pages and notes once, and it cross-references every anomaly against everything it knows. One click, one paragraph, citing the exact document and date.
30-day trial · no card · flat pricing, never % of spend
A memory for each client.
The context that explains performance lives in inboxes, Notion pages and meeting notes. Everywhere except your reporting tool. Context Hub gives it one home, per client.
Indexed on upload
Documents are chunked and embedded automatically within seconds of upload, then retrieved on demand by Flume and the narrative generator. No manual tagging, no curation chores.
External Changes timeline
URL diffs, CSV imports, manual entries, per-platform creative-change dates synced automatically. The ledger of everything that happened outside ROASt that might have moved the numbers.
One click from anomaly to answer.
When something moves, the explanation draws on everything in the store and points at the exact document and date. No guessing which change caused the dip.
-
“Explain this” on every anomaly
One click on any anomaly returns a narrative paragraph: ranked causes, relevant context, engagement signals, hedged language when confidence is low. Pasteable straight into a client reply.
-
Cites your own docs
“ROAS dipped two days after the landing-page rewrite logged on the 14th” beats “performance fluctuated.” The narrative names the source it drew from, every time.
-
The weekly brief
Monday morning, dashboard and email: pacing, top movers, anomalies, creative and keyword opportunities, competitor watch. The same causal plumbing, written for an operator triaging multiple clients.
It teaches itself. Every narrative lists the data sources it
couldn’t consult because they’re empty, with a deep link straight into
Context Hub to add them. Richer context in, richer answers out.