🎯 Valent Pilot — Strategic View

Valent is not a $5K pilot. Valent is proof of concept for the entire AIPMO product. If this works:

• First external validation that Warren/AIPMO delivers value to a real consulting firm
• Repeatable playbook — same bootstrap kit can deploy for Kodzen, Acorns, NBC Universal, Clarkston, Unify
• Reference client — Hector has "several friendly clients" for expanded pilot if internal eval succeeds
• Revenue proof point — $5K pilot → $16K/month commercial → $200K-$500K/year per engagement = Valent's 1-2 FTE savings

Tony defined the split (June 5 weekly sync):

Key insight: Warren is already beyond what most customers can understand or evaluate. The analytical tier (process-following, deliverable production) is sufficient to exceed client expectations. The efficacy tier (80/20 judgment, paradigm shifts) is our competitive moat — it's what prevents anyone from catching up by just installing OpenClaw.

From the WWTD protocol — this is the strategic positioning framework:

What the customer buys:

"AI runs your PMO process better. Produces deliverables faster, catches risks earlier, keeps everything aligned."

What the customer actually gets:

Removal of the process layer from their humans. The process IS the obstruction — it consumes cognitive budget that should go to direct perception. Warren absorbs the analytical noise (the 80% that's automatable) so consultants can focus on clear seeing (the 20% that requires human judgment).

Why this matters for Valent:

Valent's consultants live in process. Sprint planning, status reports, backlog grooming — that IS their day job. If Warren does the process work AND asks better questions than a junior PM would, Valent's senior consultants are freed to do what Hector, Nick, and RJ are actually good at: client relationships, strategic judgment, and seeing what's really happening in a program.

Why can't Valent just build this themselves?

Matt Malone (their AI principal) will evaluate this. He could theoretically set up an AI agent. Here's what he can't replicate:

• 3+ months of calibration — Standing rules forged by real failures. Evals system with shadow review, multi-judge, calibrated rubrics. 86-entry corpus from human verdicts.
• See It Cycle methodology — Not generic project management. A specific perception-action loop designed for consulting delivery, refined through live client work.
• Pareto Lens communication DNA — Every interaction questions whether the team is focused on the 20% that matters. This isn't a prompt — it's architectural.
• Progressive autonomy framework — Two-tier review (accuracy vs judgment), function-level graduation tracking, continuous recalibration. The system gets smarter with use.
• Cross-client learning — Warren learns from every engagement. The methodology improves across all deployments simultaneously.

In short: they could install the software. They can't install the judgment.

Valent is the template. Every success metric, every process refinement, every lesson learned from this pilot feeds directly into the next 6 deployments. The $5K pilot is a $1M+ pipeline investment.

Risk 1: "It's just ChatGPT with a wrapper"

If Matt concludes this: the deal is dead — not just Valent, but every technical evaluator we face. Mitigation: The See It Cycle, Pareto Lens, and Maybe Gate are not prompts — they're embedded methodology. The first deliverable must demonstrate judgment, not just formatting. Warren must ask a question that RJ hadn't considered. That's the moment.

Risk 2: First deliverable underwhelms

We get one shot at first impression. If the first sprint backlog looks like something ChatGPT would produce, we've lost the narrative. Mitigation: Front-load the methodology. First deliverable should demonstrate the See It Cycle — not just "here's your backlog" but "here's what I see as the highest-value items, here's what I'd cut, here's the risk nobody mentioned." Show the thinking, not just the artifact.

Risk 3: Scaling the bootstrap

If each client deployment requires this level of manual setup: we can't scale. Mitigation: Valent is the template. Document everything. After this pilot, the bootstrap kit + kickoff playbook should be reusable with 80% less effort for the next client. That's the real deliverable — not just Valent success, but Valent-as-template.

Risk 4: Tony's paradigm shift mid-pilot

If we switch the Valent instance to the new deterministic-outcomes paradigm mid-pilot: quality could drop. Mitigation: Client instances stay on the analytical tier. Warren experiments with the new paradigm internally. Only graduate to the new paradigm when it's proven on Warren first.

Client-Facing (SOW)

• Deliverables need minor edits, not rewrites
• Natural interaction via Teams
• Clear business insight from the experience
• First deliverable within 48h of data ingestion

Internal (Strategic)

• Bootstrap time: how fast can we go from signed SOW → productive AI partner? Target: <5 days.
• First value moment: when does RJ say "oh, that's actually useful"? Target: within first week.
• Methodology differentiation: does RJ notice Warren asks better questions than expected? Does Hector see the customization potential?
• Template reusability: can we deploy the next client in <2 days using the Valent template?
• Upsell signal: does Hector or Nick ask about Phase 2 before we bring it up?