Budget Pacer
Reallocates daily spend inside each portfolio nightly. Fits a saturation curve per campaign from your last 30+ days, then solves for the mix that maximises ROAS within your velocity caps and budget locks.
[03/07]Agents
/ the product tour
Meet your AI ops team.
Nine specialists working your account in the background. Three Optimisers act on it, three Watchers alert you to changes, two Analysts make sense of the data — plus a Generalist you can talk to in plain English. Powered by Claude where it helps. Deterministic math everywhere else.
[01]The roster
/ who does what
Nine specialists. No black boxes.
Some act on the account, some watch for changes, some explain the numbers. Each card says what powers it — because “AI” should never be a hand-wave.
Search Waste Hunter
Scans your Google search terms every Sunday. Deterministic spend and close-rate filters pick candidates; Claude ranks each one for relevance against your portfolio goal.
Creative Fatigue Analyst
Spots fatigued ads across Google RSA, PMax and Meta. Fatigue signals find the candidates, Claude judges them against your goal — and sibling-count guards make sure a campaign is never left without a working ad.
Pacing Watchdog
Flags portfolios drifting off their monthly spend pace — a daily digest when month-to-date spend runs more than 15% off the prorated target, worst offenders first.
Market Intelligence
Reads Google auction insights hourly to catch competitors entering or leaving your auctions — and logs each event so later ROAS shifts can be traced back to specific market moves.
Health Officer
Continuously grades measurement health — tracking gaps, broken UTMs, revenue divergence between ad platforms and your shop or CRM — and proposes specific fixes when something breaks.
Model Architect
Re-fits every campaign’s response curve nightly, then grades the engine’s own predictions against what actually happened. The receipts live in the Calibration tab.
Brief Author
Writes your weekly brief and daily mini-brief — plain-English prose that narrates what the team did, agent by agent. A real ops report, not a chart dump.
Ask Flume
Chat with full team access. Ask anything in plain English — Flume reads your data, proposes actions, and hands structured plans back. Your own AI tools can direct this team too — the full story lands on the Flume stop.
Honest framing. Five of these agents are deterministic math. Two
are math + Claude ranking — Claude judges candidate relevance against your
goal description; math handles the filtering and the apply. Two are Claude-led:
Brief Author writes the brief, Ask Flume drives the chat. No agent autonomously
changes budgets or targets — that’s Budget Pacer’s job, and
Budget Pacer is pure math operating inside your own caps.
[02]Keywords
/ the Search Waste Hunter at work
Wasted spend, hunted weekly.
Negative keywords and pause candidates are the highest-ROI tactical work in PPC — and the thing most accounts do worst. The Search Waste Hunter scans your search-term report and active keywords every Sunday and stacks its findings in one review modal.
Close-rate-aware, not just conversion-aware
A standalone tool flags “this term had 0 Google conversions.” ROASt flags “this term drove 8 leads that never closed in your CRM.” When close-rate context exists — HubSpot or Shopify — the suggester routes through it.
Goal-semantic intent matching
Your portfolio goal description is fed to Claude alongside each term. “Second-hand fitting request” gets flagged as off-intent because Claude knows what the portfolio is actually for — a judgment a regex can’t touch.
Pause + negative in one apply
Pause candidates and wasted-spend search terms together — one Apply button, parallel API calls, consolidated result. Sibling-count guards mean you’ll never strand an ad group with too few keywords.
Google-only at launch — Microsoft and Meta search-term sync are on the roadmap. Suggestions expire and reactivate on real signals: three strikes, two-week absence, or a 2× spend resurgence.
[03]Creative
/ the Creative Fatigue Analyst at work
Kill underperformers. Keep the winners.
Creative is the only daily lever on Meta — no keywords, automated bidding, automated audiences. The Creative Fatigue Analyst scans Google + Meta creative weekly and ships one unified review: pause Low Google assets and fatiguing Meta ads in a single apply.
Six creative sub-views
RSA assets, PMax asset groups, Shopping products, Meta ads, a Meta winning-creatives library, and a cross-platform Creative Health rollup. One nav, no platform-switching.
Fatigue judged in context
Low RSA assets, fatiguing PMax assets, Below-average Meta ads with rising frequency — each judged against your portfolio goal description by Claude, not a hard threshold.
Engine cross-confirmation
When the scanner’s fatigue finding agrees with the engine’s anomaly detector on the same campaign, the Audit tab raises a critical finding. Two systems agreeing is a much stronger signal than either alone.
Pause-only by design — ROASt won’t generate copy, duplicate ad sets, or edit audiences. Every apply uses the smallest reversible unit per platform, and sibling-count floors (RSA ≥3, PMax ≥4, Meta ≥1) are re-checked at apply time so a campaign is never stranded without an active ad.