The questions worth asking.
Questions that come up on most discovery calls. If yours isn't here, book one.
About Formaum
What does Formaum actually build?
Bespoke AI automation systems for multi-location operations. The work blends software engineering and operations engineering. Vertical: production code, schemas, services, APIs, observability. Horizontal: CRM migrations, multi-location workflows, AI agents, internal tooling. End-to-end. Audit, architect, build, deploy, optimise.
Is Formaum a team?
Solo studio. The engineer on the discovery call writes the code and ships the system. Trusted contractors join when capacity requires it. Never strangers.
Who is Formaum a good fit for?
Multi-location operations with real revenue and real operational drag. Industry-agnostic. Education, franchises, professional services, real estate, gaming, entertainment. The signal: five or more tools, manual handoffs, pipeline leaking through the gaps.
Who is Formaum NOT a good fit for?
Solo operators a no-code tool already covers. Pre-revenue startups seeking free spec work. Buyers wanting a strategy deck without a production build. Operations not yet generating enough revenue to absorb the engagement should wait.
Engagement & Pricing
What's included in the Operations Overhaul?
End-to-end engineering: audit, architecture, build, integration, pilot deployment, optimisation, training, handover. CRM migrations run 3 to 6 months depending on size, complexity, and team. The audit maps the stack and quantifies where revenue and time leak. Architecture follows. Build, integration, and pilot occupy the central sprints. Optimisation, training, handover close. You finish with a consolidated system you own, your team trained, and an operations layer engineered to run unsupervised under continuous load. Investment is scoped after the audit and discussed on the discovery call. Contracts paid upfront before each month.
How long does an engagement take?
3 to 6 months depending on size, complexity, and team. The audit anchors week 1. Architecture follows. Build, integration, and pilot occupy the central sprints. Optimisation, training, and handover close. Exact scope is set after the audit.
How is pricing structured?
Engagements are scoped against operational outcomes, not hours. Exact scope is set after the audit.
What does the discovery call cover?
No pitch. We map current operations, quantify the cost of staying outdated quarter over quarter, and test whether the Operations Overhaul fits. Operators who consolidated this layer 12 to 18 months ago now ship faster on every customer touchpoint than their direct competitors do. The structural gap widens monthly.
Process & Delivery
What's the working cadence?
Sprint-based with a defined scope and clear definition of done. Weekly check-ins. Async updates between calls. Full visibility into what ships and when. No surprises in the middle. No scope creep at the end. The schedule is the product.
Do I own the system at the end?
Yes. The system, the code, the documentation, the playbook, the credentials. No proprietary Formaum platform. No lock-in. The work is yours from day one.
How do you handle multi-location rollouts?
Pilot-first, always. Launch at one location. Run it for a week against real production data. Fix what surfaces. Roll to the rest. A single mis-mapped field in a multi-location routing engine sends inbound to the wrong location. Pilot validation prevents that at scale. Multi-location is sold upfront as a phased plan.
Tools & AI
What tools does Formaum work with?
Claude for AI. Claude Code for engineering. Trigger.dev, n8n, Make for orchestration. Supabase, PostgreSQL, BigQuery for data. GoHighLevel, HubSpot, Salesforce, Pipedrive, Monday for CRM. Twilio, Vapi, ElevenLabs for voice. Next.js, FastAPI, Python for custom apps. Tool selection is per-job, not per-vendor.
Are AI agents safe in production?
Yes, when engineered for production. Guardrails, observability, graceful degradation. Every action logged. Confidence thresholds route low-confidence decisions to human review. The system fails to a safe state when a model is unavailable. AI at scale without engineering judgment corrupts records and mis-routes inbound silently. The boring infrastructure is what separates a demo from production.
Why Claude over OpenAI or Gemini?
Claude is the default. Reasons: model quality, Claude Code and Skills tooling, data privacy posture on Team and Enterprise plans. Other models are used where the job calls for them. OpenAI for embeddings, Gemini for visual review. Per-job, not per-vendor.
Privacy & Geography
Will my data and IP be safe?
Yes. NDAs signed before any client material is shared. AI providers run on Team or Enterprise plans where applicable. Client data is not used for model training. Production credentials live in secrets managers. Audit logs cover every agent action.
Do you work with clients outside Canada?
Yes. Most engagements run remote across Canada, the US, and the UK. Time-zone overlap is built into the sprint cadence.
Can I see your previous work?
Selected work is on the Case Studies page. Names withheld under NDA. Numbers and architectures are real. Reference calls arranged for serious buyers.
No questions match that search. Try a different keyword, or book a call.
Question not covered here?
No pitch. Map your stack, quantify the cost of staying outdated, decide on the production-grade alternative.
Book a discovery call