How a SaaS company cut response times from 4 hours to 12 minutes by routing support queries through local AI.
Northstar Housing provides property management software to housing associations and local authorities. 200 employees, 800+ customers, and a support desk that handles 150-250 tickets per day. Their support team is six people. They're good at their jobs. They're also drowning.
The problem isn't volume. It's that the 70% of routine tickets block the 30% that actually need a human. A housing officer emails at 9 AM: "I can't log in." That sits in the queue behind three bug reports and a billing question. By the time someone responds at 1 PM, the housing officer has been locked out for four hours.
The cost: Average response time 4-6 hours. CSAT 68%. Two agents spend their entire day on password resets. Senior engineer spends 2 hours/day on triage. Three enterprise contracts with 1-hour SLAs — they're missing them.
They tried hiring more agents (volume grew again), building a help centre (12% usage), and Intercom's AI bot (sent sensitive data through OpenAI — procurement flagged it as a data processing risk).
Foundry runs on a Mac Studio that Northstar already had for video production. It acts as a first-responder for support tickets.
When a ticket arrives: Foundry reads it, checks the knowledge base, drafts a response for routine questions, verifies identity for account-specific actions, reproduces and routes bug reports, and passes complex issues straight to a human with a summary.
Every response is a draft. A support agent sees the draft, confirms it's right, and clicks send. For routine tickets, that takes 10-15 seconds of reading — not 5 minutes of typing.
Over time, Foundry learns from the agent's edits. If the agent corrects the same thing three times, Foundry starts drafting it that way.
Response time (routine): 4-6 hours
Tickets per agent/day: 25-30
CSAT: 68%
SLA breaches: 4-6/month
API cost: £450/month
Response time (routine): 12-15 minutes
Tickets per agent/day: 60-80
CSAT: 91%
SLA breaches: 0/month
API cost: £0 (local)
Annual impact: £35,000-45,000 in freed staff time + £5,400 in API savings + retained enterprise contracts worth £120,000/year that were at risk due to SLA breaches.
Foundry cost: £999 setup + £99/month = £2,187 first year.
"The first morning after we turned it on, I came in to 23 tickets and thought 'here we go.' Twenty minutes later I was getting a coffee. It's a different job now." Support agent
"I was skeptical. I thought it would send something stupid to a client and we'd lose a contract. But nothing goes out without my say-so. The drafts are good — usually 90-95% right. I just tidy and send." Senior support engineer
"Our enterprise clients asked us last year to stop using OpenAI for ticket processing. We couldn't find an alternative. Foundry was the answer — it's local, we own it, and their procurement teams are satisfied." Tom, MD
Works for SaaS companies with 50+ tickets/day, B2B companies with data-sensitive clients, teams where routine tickets block complex issue resolution.