Pursuit intelligence. On your infrastructure.

Your firm has the evidence.
Field Thesis turns it into a thesis your team can defend.

Turn filings, earnings calls, internal notes, and firm POV into a defensible account position — before the next executive conversation.

Runs entirely in your environment. No client data, account intelligence, or firm IP leaves your infrastructure.

Built for teams selling complex technology, consulting, and transformation services — where credibility, timing, and judgment shape the deal.

Current Field Thesis+1 new signal this week

“AI pilots are outpacing the operating model. Path-to-production controls are becoming the bottleneck.”

Strongly indicated
Evidence trail
  • 10-K · Risk factors expanded around third-party dependencies
  • Q3 earnings · Expense discipline tied to platform simplification
  • Account note · CIO piloting GenAI without production controls
Recommended move

Validate path-to-production friction with CIO / CTO team

Auditable trail · Calibrated language · Recommended moves

Behind Field Thesis

Built by people who've sat in the rooms.

Field Thesis was built by former management consultants and enterprise sellers who spent their careers preparing for C-suite pursuit meetings — not prototyping AI demos.

We wrote the Sunday-night thesis the Monday meeting deserved — what you may call a pursuit strategy, a POV, or an account stance. We watched account teams arrive over-prepared and under-positioned — months of account briefs, stakeholder maps, prior pursuit work, none of it synthesized into a view the partner could defend. We watched the firm's institutional point of view — earned across years of prior pursuits — sit outside the room when the team needed it most. We watched fluent-sounding research dissolve under the first hard question from a CFO who actually knew the business.

Field Thesis is the system we wished we had — built for the partners, managing directors, and pursuit leads who still walk into those rooms.

The Problem

Surrounded by intelligence. Short on judgment.

Enterprise teams have more account information than ever. But the judgment behind major pursuits still depends on scattered documents, individual experts, and last-minute interpretation.

01

Intelligence does not equal judgment.

Teams summarize the account but walk into the room without a position. Recapping isn't winning.

02

Right information. Wrong moment.

Signals are buried across filings, earnings calls, account notes, and prior pursuits. Firm POV and expert knowledge sit outside the moment when teams are deciding what to say.

03

Evidence strength is uneven.

Some claims are safe to assert. Others should be softened. Others should become discovery questions. Most tools blur those lines.

Why Field Thesis

A new category: pursuit intelligence.

Field Thesis does not simply summarize account information. It helps teams move from evidence to interpretation to action, so their positions are traceable, calibrated, and reusable.

01
Most tools

Conversation intelligence tells you what happened in the call.

Field Thesis

Tells your team what to say before they're in the room.

02
Most tools

Enterprise search finds the document.

Field Thesis

Tells you what your firm already thinks about it — and what's worth testing next.

03
Most tools

Account-based platforms tell you which accounts to target.

Field Thesis

Tells you what thesis to bring once you're in.

04
Most tools

AI sales coaches let you practice against generic personas.

Field Thesis

Grounds simulations in the actual stakeholder's 10-K.

05
Most tools

Generic AI produces fluent answers regardless of evidence strength.

Field Thesis

Calibrates or withholds — and every claim quotes its source.

Pursuit intelligence is what your team walks in with — not what they leave with.

The Platform

Three surfaces, one intelligence spine.

Query for strategic research, War Room for active pursuits, Simulate for stakeholder practice — all reading from the same evidence base, calibration layer, and firm POV.

01

Query

Strategic research in under a minute.

Semantic search across the firm's knowledge base, augmented with live external intelligence from ten sources. Every response ends with 2–3 testable hypotheses — and when you ask 'what's our POV on X,' Query compares the firm's prior position against current evidence side-by-side.

Where does our “governance must lead” POV land in this account today?

The firm's view holds. The CFO is reachable — but only through an operating-model wedge, not an AI thesis.

Medium-High
POV Route
Firm POVQ1 2025 whitepaper

“Governance must lead pilot expansion.”

Current evidence10-K · Q3 · Account note
  • 10-K: New third-party AI dependency language (audit committee, not risk factors)
  • Q3: CIO tied to “platform simplification,” not pilot expansion
  • Account note: 3 BU GenAI pilots, no production owner
Hypotheses to test
  • CFO will fund governance if positioned as operating model, not AI.
  • 10-K audit-committee origin is the strongest evidentiary entry.
  • CIO owns pilots; nobody owns production controls.
02

War Room

Persistent pursuit intelligence.

Per-pursuit container structured around Miller Heiman Strategic Selling. Ten intelligence sources fire in parallel and persist. Background daemon refreshes filings every six hours; macro data daily.

GIG
Global Industrial Group
10-KEarningsInvestor DayNote

AI pilots are outpacing the operating model.

StakeCIOCTOCFO
03

Simulate

Practice against the actual stakeholder.

Stakeholder simulations grounded in the actual 10-K, capex, and competitive landscape. Persistent per-persona memory across sessions. Hypothesis testing with hidden ground-truth values revealed in debrief.

CFO
CFO · Session 3 of 4

I read “platform simplification” as governance. Am I right?

CFO

“You're reading more intent into that than I put there.”

Hidden hypothesis · revealed in debrief

The Pursuit War Room

The thesis your team carries — and the system keeps current.

The War Room connects source evidence, firm knowledge, signal interpretation, confidence bands, pressure-test questions, and recommended moves. Every claim is traceable. Every inference is labeled. Every action ties back to the evidence.

GIG
Global Industrial GroupIndustrial Conglomerate · Pursuit thesis
4 sources3 signals3 questions3 moves
1

Source Evidence

10-K Risk Factors

Expanded disclosure around third-party technology dependencies and operational disruption risk.

Q3 Earnings Call

Management tied expense discipline to platform simplification and automation priorities.

Investor Presentation

Digital roadmap emphasizes core modernization, control, and efficiency.

Internal Account Note

CIO team is piloting GenAI use cases but lacks a clear production control model.

2

Signal Interpretation

Across 165 named pattern types

Control and release governance are becoming execution bottlenecks

AI pilots are moving faster than the operating model can support

Modernization language is shifting from experimentation to operating discipline

3

Current Field Thesis

The account appears to be moving from AI experimentation toward a governed path-to-production agenda. The strongest evidence points to pressure around control evidence, resilience, observability, and release governance rather than pure innovation spend.

ConfidenceMedium-High
Allowed claim language

“appears to be moving” “strongest evidence points to”

Commercial Frame

Theme
Governed speed to production
Entry
CTO / CIO operating model
Still to validate
  • ?Is AI governance a funded priority or still exploratory?
  • ?Who owns the path-to-production decision?
  • ?Does this align with prior pursuit findings?
4

Pressure-Test Questions

?Where are AI initiatives getting stuck between pilot and production?
?Which controls are slowing release velocity today?
?What would leadership need to fund a broader modernization roadmap?
5

Recommended Actions

Prepare CIO briefing on governed speed to production

Near-term

Map SDLC bottlenecks against control requirements

Near-term

Revisit prior DevOps scope through AI lens

Discovery
Confidence calibration
SupportedSuggestedRestrictedWithheld

This is how pursuit teams move from “we researched the account” to “we have a position we can defend.”

On-prem
No data egress
Verbatim
Every claim sourced
Calibrated
Four-band confidence
All majors
Industry packs

The Simulate Surface

Practice the conversation. With the buyer whose 10-K you just read.

Simulate grounds each conversation in this stakeholder's actual filings, tests your hypotheses against truth values hidden until debrief, and remembers what was said across sessions. Other AI sales coaches practice against templates.

CFO
CFO · PersonaSession 3 of 4 · Industrial Conglomerate
Hidden hypothesis · revealed in debrief
You

Your Q3 transcript mentioned 'platform simplification' three times. I read that as a governance signal. Am I right?

CFO
CFO

“You're reading more intent into a word than I put there. Finance wrote that paragraph. Don't anchor on three words.”

You

Fair. But the 10-K added new third-party AI dependency language this year. That's not Finance writing risk disclosures — that's the audit committee.

CFO
CFO

“Closer. But if you walk into this room, don't bring me another AI thesis. Bring me an operating model conversation. The AI question is downstream of whether we have a governance function that can actually own anything.”

End of Session · Debrief
Hypothesis tested

“Capex pivot is governance-led, not budget-led.”

Ground truthPartial
  • ✓Governance is the lever — CFO will engage ‘operating model’ framing
  • !‘AI governance’ framing rejected — reframe before the meeting
Memory captured for next session
  • CFO dismisses ‘platform simplification’ as evidence of intent
  • CFO will engage ‘operating model,’ not ‘AI thesis’
  • 10-K audit-committee origin: accepted as signal
  • 2026 envelope: not yet probed
01

Grounded, not generic

The persona is built from this account's actual 10-K, capex, competitive landscape, and organizational disclosures — not a role-archetype template. The persona's answers are bounded by what the disclosures actually say.

02

Memory that persists across sessions

The system remembers what was said, what was committed to, and what was probed — per persona, per pursuit. The CFO's memory doesn't bleed into the CIO's. Session three knows what happened in session one.

03

Hypotheses with hidden ground truth

When you select hypotheses for a session, the engine assigns truth values — hidden during play, revealed in debrief. This is hypothesis validation through simulated discovery, not roleplay theater.

04

Same discipline as the rest of the platform

The persona obeys the evidence rules. If the supporting language doesn't exist in the filings, the persona doesn't fabricate it — it hedges instead of guessing.

This is how pursuit teams practice the question they're actually going to be asked — not the question a generic AI thinks they should be.

The Discipline

Built for teams that cannot afford confident-sounding guesses.

Field Thesis separates source evidence from interpretation, calibrates language to evidence strength, and turns weak signals into questions instead of unsupported conclusions. The discipline is enforced in the engineering — not just claimed in the marketing.

Same signal. Different language.

One underlying pattern. Four levels of what the team is allowed to say in the room.

01Supported claim

The account is prioritizing operational resilience.

Multiple corroborating sources. Safe to assert directly.

02Suggested claim

The account may be elevating operational resilience as a board-level concern.

Emerging pattern. Use calibrated, hedged language.

03Restricted inference

Where is resilience creating execution pressure today?

Indirect signal. Convert to a discovery question.

04No claim

No claim generated.

Evidence is too weak, stale, or procedural. Field Thesis withholds instead of fabricating.

How it's enforced.

Three engineering controls sit underneath every output — quietly rejecting the claims that don't hold up.

01

Source-locked claims

Every claim must quote an exact phrase from a source document — verbatim. If the model can't find one, the claim is rejected. Nothing gets paraphrased into existence.

02

Claim auditor

Before any output reaches the team, a separate auditor checks every claim against its calibration band. Overclaim and underuse are both caught — a claim that uses a verb stronger than its evidence supports is blocked, and a confirmed claim buried under hedged language is flagged just as hard.

03

Withhold over guess

When evidence is too weak, the system returns nothing rather than fabricating language. Abstention is treated as a correct outcome.

Beyond Enterprise Pursuits

The same discipline travels.

Substring validation, calibration bands, and hypothesis ground-truthing are already the professional standards for research teams. Field Thesis enforces them in the engineering — wherever the thesis lives.

01

Hedge fund research

Fundamental long/short, event-driven, special situations. Building positions, defending in IC, prepping for management calls — the workflow Simulate was built for.

02

PE deal teams

Deal associates, VPs, principals. IC memos, prior-cycle thesis defense, management meetings — held in War Room, kept current by the system.

03

Activist research

Campaign building, proxy preparation, public letters. Theses that survive SEC scrutiny and adversarial board engagement. Every claim source-locked because the alternative is a retraction.

04

Equity research

Sell-side and buy-side analysts maintaining coverage. Calibration bands and sourcing standards are already required by professional standards. Field Thesis enforces them in the software.

One platform. Same engineering. Different rooms.

Bring this to your next pursuit.

The Transformation

What changes when intelligence becomes a position.

Before Field Thesis

Raw FragmentsUncalibrated
  • 01
    10-K

    Cloud line items disclosed at higher granularity than prior year.

  • 02
    Q3 Earnings

    Management language shifted from 'cost discipline' to 'unit economics.'

  • 03
    Analyst Day

    Technology spend framed against revenue per workload, not headcount.

  • 04
    Account note

    CIO referenced 'workload accountability' as the operating model gap.

  • 05
    Prior pursuit

    FinOps assessment from 18 months ago stalled at the BU level.

  • 06
    Firm POV

    Chargeback discipline is the lever procurement-led approaches keep missing.

No shared interpretation. No calibrated claim. No next move.

With Field Thesis

Field Thesis OutputMedium-High

“Cloud spend is outpacing workload value. The lever isn't the hyperscaler contract — it's the chargeback discipline the operating model doesn't enforce.”

Evidence trail
10-KQ3 EarningsAnalyst DayAccount note
Signal interpretation
  • Cloud line items are growing faster than the businesses they support.
  • Margin language has shifted from cost discipline to unit economics.
  • Internal references to 'workload ownership' are appearing in reorg signals and senior hiring posts.
Pressure-test questions
  • ?Is finance modeling cloud as fixed cost or variable today?
  • ?Where does workload ownership actually sit in the current operating model?
  • ?Has any BU successfully run a chargeback regime that finance accepted?
Recommended moves

  • Brief CFO on workload chargeback as the unlock

    Near-term

  • Map prior FinOps scope against current waste

    Near-term

  • Discovery with most-instrumented BU on unit economics

    Discovery

Field Thesis does not create more reading. It creates a position your team can defend, pressure-test, and act on.

Get Started

Show up with a thesis,
not a recap.

Move from scattered intelligence to a thesis your team can defend, pressure-test, and turn into the next strategic move.

Design partner pilots run 6–8 weeks. One active pursuit team. By invitation.

Built for teams selling complex technology, consulting, and transformation services — where credibility, timing, and judgment shape the deal.

What your team walks in withPer account
Headline outputMedium-High

One defensible account thesis.

Grounded in evidence your team can inspect, with calibrated language for the room.

  • 014
    Supporting sourcesLinked to exact citations
  • 023
    Commercial signalsPatterns that may change the pursuit
  • 033
    Pressure-test questionsGenerated from thesis uncertainty
  • 043
    Recommended movesTied to commercial interpretation
Every claim calibrated
Supported Suggested Restricted Withheld