Tier 1 · Sprint
From zero to production agents in eight weeks.
One to three Agentforce use cases, built right the first time — guardrails, evaluation, observability, and team enablement included.
The problem
You want production agents — not a nine-month consulting programme.
You’ve bought Agentforce, or you’re about to. You want it live, in production, doing real work — not in a multi-phase consulting programme that bills for nine months and ships something you can’t operate. You want a scoped first deployment that proves the model, with the evaluation and observability rigour built in from the start.
The Launch Sprint is built for exactly that buyer.
What you’ll have at the end
- One to three Agentforce use cases live in production, against agreed KPIs
- A data foundation for those use cases (scoped, not enterprise-wide)
- An evaluation framework you keep — the same harness customers normally pay separately for
- Production observability you can run yourself
- Guardrails, escalation, and compliance posture aligned with EU AI Act (IE / UK) or KSA NLA (ME)
- Your internal team enabled to extend and run the work
Scope
A clear line between what’s in and what isn’t.
Included
- Use case scoping and design (up to three at kick-off)
- Agent build, prompt, and topology
- Data foundation work for the agreed use cases
- Evaluation framework
- Observability and monitoring
- Escalation, guardrails, and audit trail
- Go-live support
- Knowledge transfer
Not included
- Use cases beyond the agreed scope at kick-off
- Enterprise-wide data platform work (routes to Data Cloud Quick-Start)
- Ongoing operation post-go-live (routes to Managed Agentforce)
How it works
The same shape as the Performance Sprint.
Weeks 1–2 — Scope & design
Use case scoping and design, up to three at kick-off.
Weeks 3–5 — Build
Agent build, prompt and topology, and the data foundation for the agreed use cases.
Weeks 6–7 — Test & harden
Evaluation framework runs against every change. Observability stood up. Guardrails, escalation, and audit trail hardened.
Weeks 7–8 — Go-live & stabilise
Go-live support and stabilisation. Daily working sessions with your team. Architect-led throughout.
Same shape as the Performance Sprint. Engagements run six to ten weeks.
Who this is for
- Net-new Agentforce buyers in mid-market and enterprise
- Organisations with an Agentforce-aware Salesforce AE pushing for a credible first deployment partner
- Teams that want production agents, not a pilot deck
How we deliver
Same team structure as Performance Sprint: senior architect lead, agent engineer, part-time data engineer, principal at gate. Senior-only.
Frequently asked questions
What’s the difference between this and the Performance Sprint?
Launch builds your first agents. Performance fixes agents you already have. If you don’t yet have anything in production, you want this one.
Three use cases sounds optimistic. Is it realistic?
Yes, when the use cases are scoped properly. Three small, well-bounded use cases ship together cleanly. Three large transformational ones don’t — and we’ll tell you so at scoping rather than discover it at week six.
Can we extend mid-sprint?
Through explicit change control, yes. Not by drift.
Will you charge separately for the evaluation framework?
No. The eval harness, observability setup, and guardrail design are included. They’re the difference between a pilot and production, and they shouldn’t be optional.
What if the engagement needs to extend?
Through explicit change control, with an explicit conversation and an explicit revised commitment. Not by drift. The fixed-price discipline is a feature, not a constraint.
How does payment work?
Payment is in three stages for sprints: 30% on signature, 40% at a mid-engagement gate, 30% on completion. The gate is genuine — if we haven’t earned the milestone, we don’t invoice it. For Diagnostics and Advisory, payment is 50% on signature, 50% on delivery.