10 production systems across growth, ops, and intelligence.
A global storage marketplace with 100+ locations needed AI embedded across the entire business. Not one system — ten. All in production. All still running.
Growth was running on manual processes, ops was flying blind, and intelligence was a spreadsheet.
StowFast operates a global storage marketplace with 100+ locations. The business was growing, but the systems weren't keeping pace.
Outreach to new partners (museums, councils, tourist sites) was manual: open Gmail, hand-write an email, copy to the CRM, set a mental reminder to follow up. Reps spent 7+ hours a week on logging and sequencing before writing a single email. Follow-ups were dropped. Replies sat in the inbox for days.
Expansion decisions — which cities to open next — were made on gut feel and a partial spreadsheet. Each site-selection debate burned 1–2 days of analyst time pulling ad-hoc data.
Review management across 100 locations was tracked in a shared sheet. Competitor pricing was checked manually. Google Business Profile posts were skipped for most locations. Link building was a monthly two-day exercise in browsing directories.
Every department had the same shape of problem: a human doing repetitive work that a system should handle, with no centralised intelligence layer connecting the data.
Ten systems. Each one solves a specific problem. Together, they run the growth operation.
MCP server connecting Claude directly to Gmail + Attio. Draft, send, log, and sequence outreach from a single conversation. 3-step follow-ups with reply detection, bank holiday awareness, and human-paced sending (5–15/hour with jitter).
FastAPI platform scoring 100 UK cities by market opportunity. Compares search demand, POI density, transport hubs, and visitor stats against actual revenue. Surfaces underperformers = biggest ROI for new stores.
Next.js dashboard tracking reviews, velocity, and competitor snapshots across all locations. Gap analysis generates review campaigns automatically. Bulk-launch across every location in one click.
Node/Express + Playwright server scraping 4 competitors across 12 locations every 10 minutes. 30-day history, price alerts, JSON API. 50k datapoints/month.
6-stage pipeline: ingest events (Ticketmaster, Eventbrite, AI research) → dedupe → match to nearest store → generate multi-language copy → publish. Designed for 100+ locations.
Playwright scraper + Anthropic managed agent. Scrapes directories, verifies domains via SEMrush, matches to position 2–3 keyword targets. April campaign: 29 links, $2,927, targeting 67k keyword volume.
CLI tool: reads Google Ads export → Playwright scrapes landing pages → Claude analyses keyword-to-page relevance → sorted worst-first report. One account audit in hours, not a week.
300 reviews per batch across 20 locations in 20 minutes. Keyword-optimised, tone-varied, compliance-validated.
Discovery, rank tracking, upvote campaigns, AI comment generation with approve/edit/reject workflow.
Self-hosted Kuma instance tracking all production systems.
The outreach team reclaimed 7 hours a week. Follow-up drop rate went from 30% to zero. Reply-to-CRM latency dropped from hours to under 5 minutes.
Expansion debates that burned 1–2 analyst days now take under an hour — open the scorecard, read the city ranking, make the decision. The platform has evaluated 100 UK cities with percentile-ranked opportunity scores, replacing a partial spreadsheet that covered 10.
Review ops went from 6 hours a week to 30 minutes. Competitor price changes are caught within 10 minutes instead of days. Google Business Profile posts now cover all locations instead of the 20% that a VA could get to.
The estimated annual value across all ten systems — combining time reclaimed, ad efficiency, revenue protection, and expansion intelligence — is £140–220k. The systems cost a fraction of that to build, and the marginal running cost is near zero.
What would 10 systems do for your business?
Start with an audit. In two weeks we'll map your workflows, find the highest-leverage opportunities, and tell you exactly what to build first.