FAQ
How is this not just another GPT wrapper?
We orchestrate 800+ specialized instructions across multiple models, each handling different aspects of strategic foresight—causal mapping, probability weighting, signal detection. GPT is one node in a complex graph. Our moat is 36 months of refinement on these prompt chains and evaluation metrics that can't be replicated.
What's your technical moat beyond prompt engineering?
Three layers:
(1) Compounding knowledge graph—every client scenario improves predictions for the next,
(2) Proprietary evaluation metrics for scenario plausibility,
(3) Real-time signal ingestion pipeline that updates probabilities nightly.
The switching cost is high once VALIS outputs drive board meeting cadence.
Walk me through unit economics at scale.
- Base fee of $1.2M a year for access to the scenario planning tool
- Setup takes a couple of weeks for a VALIS strategist
- Each VALIS Strategist can handle 6 clients a year
- Aggressive growth of sales team to minimum 30 by Year 5
- At scale, 70%+ gross margins are achievable
Why subscription vs usage-based pricing?
Strategy executives budget annually, not per query. $100k/month is less than two McKinsey partner weeks (~$120k).
We may also offer usage-based options; the base product can be expanded upon with a long list of quality-of-life features, such as the ability for VALIS to actively read your emails and deliver insights before you open them.
Who's the buyer and what's the cycle?
We are targeting Tier 1 Lighthouses in the Deep Tech industry, where we have unique c-suite access through our partners.
Economic buyer: Chief Strategy/Innovation Officer.
Budget holder: CFO.
Close: 4-6 months.
We compress decision time by pricing at "rounding error" level for strategic initiatives.
VALIS will deliver a package of unique insights to prospective clients as a first step in the cycle.
How capital efficient can this really be?
Deck shows path to $132M revenue with <10 engineers and <12 strategists.
We expect heavier lift up front as we develop the tools for our clients to do full VALIS runs themselves.
Aside from a large sales team, we stay small and nimble. There is no reason why we would not be able to maintain the 70% gross margin.
What's the real TAM beyond consulting?
416 frontier tech companies generating $101B ARR today.
Assume 5% of strategy/R&D budgets shift to continuous foresight = $5B addressable.
Plus sovereign AI and defense markets. Every company making existential bets needs this.
Exit strategy?
Strategic buyers: McKinsey QuantumBlack, Palantir, BAE Digital Intelligence. Comparable exits at 10-15x forward revenue (Recorded Future, Primer.ai). But honestly, this could be the platform that owns strategic intelligence for the next decade.
What keeps you up at night?
Comparison to data-driven scenario planning tools; some companies struggle to see why data failed Nokia, and how data was not responsible for Apple's success (under Steve Jobs).
Long sales pipelines, ability to maintain momentum and morale.
Can I customize how VALIS thinks?
Yes. Every framework is tunable—adjust how heavily you weight different signals, which uncertainties matter most, or upload your proprietary models.
How does VALIS connect to my existing tools?
Clean API endpoints push live probabilities into whatever you use—Snowflake, Tableau, Palantir, Power BI. Your team sees VALIS insights wherever they already work. No new dashboards to learn.
What if my board can't grasp a wild-but-possible scenario?
VALIS includes "Narrative Bridges"—step-by-step stories showing exactly how we get from today to that future. Turns "that could never happen" into "oh, I see how that could happen."
How long until my team is productive?
A week or two of setup. If your team can write a strategic brief, they can use VALIS. No data science degree required.
What's coming next?
Real-time monitoring trials with pilot clients, video and satellite imagery ingestion, adversarial AI that hunts for your blind spots, and multiple ways to rank and filter scenarios. Roadmap is shaped by pilot partner needs.
Can I see under the hood?
Yes, we can run a thesis specific to you and then show the architecture and the impact it has on select components.