Folio (Folio AI; operating entity VERSO.AI SAS)

June 15, 2026

Public disclaimer: This is an independent public teardown based solely on publicly available information retrieved on 2026-06-15. It is research support, not investment, legal, tax, or financial advice. Company-provided statements are labeled as claims, not verified facts. Folio did not participate in or review this report. Errors and omissions are possible; final decisions remain with the reader.

The person who wrote Hugging Face's smolagents is now shipping an AI agent that edits native PowerPoint and Google Slides decks, and that authorship is the single most independently verifiable thing about the company. Folio runs out of a French software house incorporated in March 2026, lists on Microsoft AppSource, and pitches consulting teams a credit-priced editing agent it benchmarks against the incumbents on its own scorecard. The builder's track record checks out on the public record; the paying customers, the retention, and the benchmark wins do not. We put 57 public sources through five independent partner lenses in a single day, every claim sourced and no founder narrative taken on faith. What is left is one question about whether a rare engineering signal can survive contact with real demand and the platforms sitting above it. The teardown answers it. Read it before you take the founder call.

What The Company Does

Folio presents itself as an artificial intelligence agent for creating and editing native PowerPoint and Google Slides decks, with a stated focus on consulting and strategy teams that need to revise existing branded decks rather than generate one-off presentation images (Folio homepage; Folio help center). The PowerPoint distribution is independently visible through Microsoft AppSource, where the listing appears as "Folio AI by Verso.ai" with one public rating; the Google Slides channel is still needs founder validation because no public Workspace Marketplace listing surfaced.

The business model is credit-based self-serve software. Folio pricing lists a free Starter tier, paid Plus and Pro tiers, and higher Enterprise and Ultimate credit tiers up to $499 per month, with overage rates. Those prices are company claim evidence for list pricing, not realized average revenue per account.

The public record identifies VERSO.AI as the French SAS behind the product, incorporated on 5 Mar 2026 with EUR1,000 share capital and non-employer status (French government registry). Founder evidence is unusually asymmetric: Aymeric Roucher's authorship of Hugging Face smolagents is independently corroborated by GitHub, while Adrien Urlacher's public profile still shows a Director of Product role at Aircall (Adrien Urlacher LinkedIn).

Key Takeaways

Investor reaction
Folio has one rare, independently corroborated technical founder signal on a sharp deck-editing wedge, but the public record contains no independent paying-customer, revenue, retention, or active-org evidence, so the right answer is a gated call rather than a check.
Current verdict
hold with medium confidence.
Why investors might lean in
The author of Hugging Face smolagents is building in an agentic workflow where native file fidelity matters, the product is live in Microsoft AppSource, and category timing is real even if execution risk is high.
Why investors might pull back
The "only agent for consulting" and benchmark-superiority claims are company claim evidence, direct substitutes such as Plus AI and Gamma address the same buyer, and value capture depends on Microsoft, Google, and model providers that also control distribution or core inputs.
Highest-leverage fix
Produce one independent paying-retention cohort from real consulting teams and pair it with a blinded third-party rerun of Folio's editing-fidelity benchmark on held-out enterprise templates.
Best next move
Take one time-boxed founder call only if the founders pre-supply adoption proof, benchmark-rerun materials, engineering-bench evidence, marketplace analytics, and a value-capture brief.

What Makes This Potentially Fundable

The fundable version of Folio is not "another deck generator." It is a native-file editing layer for branded enterprise decks where the hard part is faithful structural manipulation of messy PowerPoint and Slides files, not making attractive first drafts. If Folio becomes the system of record for consulting teams' template grammar, brand rules, stored instructions, and review workflow, the product could create switching costs that a broad suite assistant does not prioritize.

That upside remains possible because Aymeric Roucher's agent-framework authorship is a real public fact, not a founder biography repeated as proof (GitHub; Aymeric Roucher GitHub). It is also stage-appropriate: the French entity is newly incorporated with only EUR1,000 registered capital, so a seed investor could plausibly be early if the cap table is clean (French government registry).

The gap is equally clear. Public evidence proves a shipped AppSource listing, company-authored benchmark materials, and a rare founder signal; it does not prove paying demand, retention, benchmark superiority, a second engineering owner, Google distribution, or durable value capture against Microsoft Copilot and Google Gemini in Slides.

The Four Holes To Close Before Fundraising

Hole	Investor Fear	What To Bring
No independent adoption proof	A shipped add-in with no outside-the-room paying retention may be a demo, not a company.	Billing cohorts, paid-seat counts, net revenue retention, and two or three contactable consulting-team references.
Differentiation is self-authored	Folio's PrezEval and connector-moat claims may be overfit to a company-run benchmark.	A blinded third-party rerun on held-out enterprise templates, plus an independent technical review of the connector layer.
Value capture may belong to the platform owners	Microsoft and Google control the core surfaces and can bundle a good-enough native agent.	Win-loss evidence versus Copilot and Gemini, proof of pricing power, and switching-cost evidence that survives native-agent improvement.
Team and runway are not legible	The public record shows a non-employer SAS, a thin public code footprint, and a second founder with a concurrent Aircall role.	Org chart, full-time commitments, equity split, cap table, cash, burn, and named engineers on the slide harness.

Decision Snapshot

One-sentence company description: Folio is a very early AI slide-editing agent for consulting and strategy teams, live on PowerPoint through AppSource and still unproven on paid adoption.
Screen: hold.
Confidence: medium.
Suggested next action: Run one time-boxed founder call gated on a pre-call proof packet.
Why this matters now: Agentic productivity is moving into office workflows, but the platform owners are also the channel and the strongest competitors.
Investor-readiness diagnosis: The founder signal is unusually strong; the evidence packet is not yet underwritable.
Best founder use of this report: Turn self-authored claims into third-party proof before raising.

IC Disagreement Map

The partners reviewed the same evidence dossier independently. The Power-law partner tests whether the outcome can return the fund; the Prepared-mind partner tests category timing and structure; the Founder-jockey partner tests team fit; the Risk-reduction partner tests fatal flaws; and the Long-horizon partner tests compounding, incentives, and follow-on logic.

Power-law partner
holdmedium confidence
The rare founder asset keeps the option alive, but the currently modeled consulting wedge is too small and there is no independent pull to extrapolate.
Would flip on
Independent paying retention and adjacent-segment adoption would move them to pursue; proof that the wedge is confined to a narrow consulting SAM would move them toward pass.
Prepared-mind partner
holdmedium confidence
The timing and editing-fidelity insight are real enough for a call, but defensibility and adoption are both self-authored today.
Would flip on
A blinded editing-fidelity win plus a retained paying team would move them to pursue; parity with native agents or no retained usage would move them toward pass.
Founder-jockey partner
holdmedium confidence
Aymeric's agent-layer fit is independently evidenced, but the hardest build has no public bench beyond him and Adrien's Folio time is unresolved.
Would flip on
Evidence of at least two engineers on the harness and Adrien full-time would move them to pursue; confirmed sole-engineer dependency with a part-time cofounder would move them to pass.
Risk-reduction partner
holdmedium confidence
No single fatal flaw is confirmed, but zero independent demand and triple-gatekeeper dependence put the deal at the pass-leaning edge.
Would flip on
Pricing power and retention that persist after a native agent would move them off the pass case; no adoption or no credible capture answer would move them to pass.
Long-horizon partner
holdmedium confidence
The entry could fit a foundational check, but no compounding asset, cap table, runway, or durable capture mechanism is evidenced.
Would flip on
Renewal driven by reused per-account assets, a clean cap table, and full-time founders would move them to pursue; no accumulating asset or no runway would move them to pass.

No partner changed vote or confidence after the follow-up evidence checks; all five confirmed hold / medium. The identical vote is evidential rather than social because the partners reached different reasons but converged on the same two unresolved gates: independent paying adoption and durable value capture.

Scenario Range

Scenario	What The Company Looks Like In 3-5 Years	Falsifiable Trigger To Watch	Earliest Evidence
Strikeout	Folio becomes a competent add-in with a credible builder but never crosses from trial usage into retained team adoption, then Microsoft or Google bundles a good-enough native editing agent.	Months after launch, founders still cannot show a logo-independent paying-retention cohort, and a native agent reaches parity on the complex-edit use case.	The public warning sign already exists: one AppSource rating, low public community signal, and no independent customer proof.
Base	Folio becomes a useful specialist for a small number of consulting and strategy teams, with real fidelity but a bounded market and limited pricing power against suite bundles.	Paid consulting teams appear, but adjacent-segment adoption and expansion pricing do not.	A founder-call packet shows real customers but only consulting-only usage and no evidence of broader workflow pull.
Home run	Folio becomes the default agentic editing layer for branded decks, compounding per-account templates, instructions, and review workflows across deck-heavy knowledge work.	A blinded third-party rerun shows a material edit-fidelity win, retained teams renew because of accumulated per-account assets, and adjacent deck-heavy roles adopt before platform parity arrives.	Logo-independent net revenue retention above 100 percent, marketplace active-org growth, and a technical validation that survives held-out enterprise templates.

What Investors Will Test

What We Looked For	Current Read	Investor Implication	Founder Prep Priority
Fund-returning path	The broad presentation-software category is large enough to care about, but the current consulting wedge is narrow and SOM is unavailable.	Investors will not underwrite "PowerPoint is huge" without a sequence from wedge to compounding platform.	Prepare a bottom-up expansion path with retained usage outside the initial consulting wedge.
Outside-the-room evidence	Product, pricing, benchmark, and moat claims mostly come from Folio-controlled surfaces; AppSource confirms shipping but not adoption.	A beautiful demo will not substitute for customer-controlled proof.	Bring billing cohorts, references, and marketplace analytics.
Prepared category insight	Editing existing native decks is a credible non-obvious wedge; whether the harness compounds is unresolved.	Investors will test whether this is a durable layer or a feature the suite owners absorb.	Prepare a technical and customer case for why native agents do not erase the value.
Jockey coverage	Aymeric's agent-layer ability is evidenced; the team coverage for rendering fidelity and GTM bandwidth is not.	The founder strength earns a call but does not cover every hardest-build risk.	Show who builds the harness, who sells, and who is full-time.
Incentive alignment	Microsoft, Google, and LLM providers are essential suppliers or channels and are not neutral stakeholders.	Value creation and value capture are different questions.	Show pricing power, switching costs, and provider cost durability.

Claim Reconciliation: Inconsistencies Investors Will Catch

Claim Or Metric	Where It Appears	Conflicting / Unreconciled Versions	Why Investors Flag It	How To Reconcile
Brand and entity name	Folio homepage Folio LinkedIn Microsoft AppSource French government registry	Consumer brand is Folio; legal entity is VERSO.AI; AppSource publisher remains Verso.ai.	Enterprise buyers reconcile legal name, publisher, contracts, and data-processing paperwork.	Provide current registry extract, publisher migration plan, and contracting entity statement.
Google Slides availability	Folio homepage Folio help center	Public Google Workspace Marketplace checks did not surface a live listing.	An add-in company overstating a channel weakens trust in every self-reported metric.	Provide the live Workspace Marketplace URL and OAuth or CASA verification status.
Enterprise readiness	Folio homepage Folio pricing	Homepage advertises single sign-on and multi-factor authentication; pricing gates those controls to the $499 tier.	The stated consulting buyer often requires these controls before procurement.	State which controls are live, available by plan, or roadmap.
Consulting-only positioning	Folio homepage Plus AI Gamma	Folio says only AI agent for consulting teams; direct and adjacent alternatives address the same buyer.	Easily contradicted category claims make the benchmark and moat story harder to trust.	Replace "only" with a narrow tested claim about native editing fidelity.
Free-credit amount	Folio pricing Folio Copilot comparison	The pricing page and older comparison copy do not align on the free-credit grant.	Minor stale-copy issues matter when pricing is credit-based.	Standardize all pricing and credit references.
Adrien Urlacher role	Adrien Urlacher LinkedIn French government registry	Public profile shows Aircall Director of Product while entity records indicate a legal officer role at VERSO.AI.	Investors will test founder commitment and GTM coverage.	Provide time allocation, equity, salary status, and Aircall transition details.

Diligence Findings: Issues To Fix Before You Raise

1 deal-breaker4 high2 medium

No independent proof of paying retention is public.
deal-breaker
Commercial diligence
Fix
Prepare cohort exports and reference permissions before investor outreach.
The core performance benchmark is self-authored.
high
Technical diligence
Fix
Commission a blinded rerun and publish enough method detail for investors to trust it.
Value capture is exposed to platform owners and LLM providers.
high
Commercial and technical diligence
Fix
Bring win-loss, pricing, gross-margin, and switching-cost evidence.
Google Slides distribution is not independently verified.
high
Product diligence
Fix
Provide live listing URL and verification status.
Team coverage is thin in the public record.
high
Team diligence
Fix
Show the harness owner map, full-time commitments, and hiring plan.
Enterprise-readiness claims are ahead of public trust evidence.
medium
Legal and security diligence
Fix
Reconcile plan gating and provide procurement artifacts.
Cap table, runway, and funding history are private.
medium
Financial diligence
Fix
Provide cap table, cash, burn, and prior instruments.

Data Room Readiness Checklist

Procurement-readiness score: not ready.
Rationale: Folio targets enterprise-style consulting workflows, but public evidence does not show a SOC 2 or ISO attestation, EU data residency, live security documentation, or procurement-ready controls beyond policy pages.

Data Room Area	What Investors Expect	Current Read	Status	Priority
Company overview & corporate	One-pager, incorporation, good standing, structure	French government registry supports incorporation and basic entity facts; current registry extract is still needed.	`partial`	High
Financials	Monthly P&L (18-24 mo), 3-yr model with assumptions, burn/runway	No public revenue, cash, burn, or runway evidence was found.	`missing`	Critical
Cap table & funding history	Clean cap table, prior rounds, SAFEs/notes, 409A	No public funding profile for the entity surfaced; name-collision profiles should not be used as evidence.	`missing`	High
Legal & IP	Bylaws, board consents, IP assignments, material contracts	Public terms exist at Folio terms; IP assignment and contracting materials are private.	`partial`	Medium-high
Product & technology	Roadmap, architecture overview, security/compliance docs	Product claims and benchmark repo are public; production connector details and trust artifacts are missing.	`partial`	Critical
Team	Org chart, key employment/advisor agreements, vesting	Founder identities are public; commitment, payroll, vesting, and bench are not.	`partial`	High
Customers & traction	Retention cohorts, ARR/MRR bridge, sales funnel, references	No independent public customer or paid-usage evidence found.	`missing`	Critical
Market & competition	TAM/SAM/SOM, competitive landscape, pricing	TAM and SAM have modeled ranges; SOM is unavailable; competitor map is public-source-supported.	`partial`	High

First-Call Agenda For The Startup

Time	Topic	Founder Goal	Evidence To Bring
0-10 min	Adoption reality	Prove the product is paid and retained, not just listed.	Billing cohort, active organizations, paid seats, net revenue retention, reference permissions.
10-20 min	Differentiation under blinding	Show that Folio wins on messy native deck edits outside founder-chosen examples.	Benchmark rerun plan, held-out templates, raw scores, technical reviewer access.
20-30 min	Value capture versus the bundle	Explain why customers pay even when Copilot or Gemini improves.	Win-loss notes, pricing tests, switching-cost evidence, renewal reasons.
30-40 min	Team and build coverage	Prove the hardest build is not a one-person dependency.	Org chart, contributor history, founder commitment, hiring plan.
40-50 min	Distribution and procurement	Clarify AppSource, Google, security, and EU readiness.	Marketplace dashboards, live Google listing, DPA, subprocessor and security packet.
50-60 min	Financing and next milestone	Show that a first check can buy a real option with symmetric gates.	Cap table, cash, burn, raise plan, milestone budget.

Business Model And Revenue Signals

Free through $499 per month credit tiers
Company pricing pagehigh
SourceFolio pricing
List pricing only; no public ARPU or plan mix.
Microsoft AppSource listing
source-backed factPlatform marketplacehigh
SourceMicrosoft AppSource
Confirms distribution, not paid conversion.
Enterprise features advertised
company claim with partial contradictionCompany pageshigh
SourceFolio homepage Folio pricing
Single sign-on and multi-factor authentication appear gated to the top plan.
LLM and subprocessor dependence
source-backed factCompany privacy policyhigh
SourceFolio privacy policy
Unit economics and compliance need private detail.

Microsoft AppSource, Product Hunt, G2, Trustpilot, Capterra, Google Search, LinkedIn Jobs, Indeed, and Glassdoor were checked and showed no independent revenue, monthly recurring revenue, annual recurring revenue, or retention evidence. These absences define the decisive commercial gap but do not prove zero paying users.

Revenue or profit is not publicly available. Any scenario involving MRR, ARR, plan mix, gross margin, or retention remains a diligence request rather than a fact.

Revenue Quality Checklist

Before this can move from hold to pursue, the founders should be ready to show MRR and ARR by paying account, paid seats, plan mix, expansion, gross revenue retention, net revenue retention, logo churn, seat churn, credit consumption, LLM cost of goods sold, gross margin, customer acquisition cost, payback, and concentration. These figures must tie across billing exports, contracts, the financial model, and source documents.

Funding And Ownership Context

Item	Public Read	Evidence Label	Diligence Request
Total raised / rounds	Not publicly disclosed; registered share capital is EUR1,000.	`unknown` for raised capital; `source-backed fact` for registered capital	Cap table, prior instruments, and current cash.
Instruments (equity / SAFE / notes)	Not public.	`unknown`	Instrument list and legal documents.
Investors	Not public.	`unknown`	Investor list and pro-rata rights.
Post-money valuation	Not public.	`unknown`	Proposed round terms and cap table model.
Cap table / ownership	Not public.	`unknown`	Founder ownership, option pool, and any advisors or angels.
Burn & runway	Not public.	`unknown`	Monthly burn, cash balance, and runway.
Next-round plan	Not public.	`unknown`	Raise size, use of proceeds, milestones, and reserve plan.

Crunchbase and PitchBook were checked and showed no disclosed funding profile for VERSO.AI beyond registered share capital. This absence carries medium weight because pre-seed companies often raise privately, but investors still need cap-table confirmation.

Founder And Team

Aymeric Roucher
Co-founder and President; independently corroborated as an author of Hugging Face smolagents.
Aymeric Roucher LinkedIn Aymeric Roucher GitHub Hugging Face smolagents
Evidencesource-backed fact for authorship; company claim for title details
Confidencehigh
Adrien Urlacher
Co-founder and legal Directeur général; public profile also shows Director of Product at Aircall.
Adrien Urlacher LinkedIn French government registry
Evidencecompany claim for current commitment; source-backed fact for entity context
Confidencemedium
VERSO.AI
Operating French SAS behind Folio.
French government registry
Evidencesource-backed fact
Confidencehigh
Folio-AI-Labs
Public GitHub organization with benchmark materials and thin public repo footprint.
Folio-AI-Labs repositories PrezEvalPublic GitHub
Evidencesource-backed fact for public repo footprint; company claim for benchmark implications
Confidencemedium
Folio company page
Public company footprint shows early-stage scale.
Folio LinkedIn
Evidencesource-backed fact for public profile fields
Confidencehigh

Founder Competency Coverage

Domain depth
Aymeric Roucher
claimed
Adrien Urlacher
claimed
What The Evidence Is
Folio's public narrative targets consulting decks; independent buyer proof is missing.
Technical build capability
Aymeric Roucher
evidenced
Adrien Urlacher
unknown
What The Evidence Is
Aymeric's smolagents authorship is corroborated by GitHub; Adrien's technical build role is not evidenced.
Product
Aymeric Roucher
claimed
Adrien Urlacher
claimed
What The Evidence Is
Folio homepage and Product Hunt show product narrative; role allocation is not independently verified.
GTM / sales
Aymeric Roucher
unknown
Adrien Urlacher
claimed
What The Evidence Is
Adrien's public profile supports product leadership background, but Folio-specific GTM commitment is not verified.
Leadership / hiring
Aymeric Roucher
unknown
Adrien Urlacher
unknown
What The Evidence Is
The entity is non-employer in the registry and the public company footprint is thin.
Fundraising history
Aymeric Roucher
unknown
Adrien Urlacher
unknown
What The Evidence Is
Public financing history for VERSO.AI is not disclosed.
Prior founding outcomes
Aymeric Roucher
unknown
Adrien Urlacher
unknown
What The Evidence Is
Prior founding outcomes are not visible in the public record reviewed.

Reliable OOXML and Slides editing fidelity across messy enterprise templates
High
Team coverage today
Agent orchestration is evidenced for Aymeric, but rendering-fidelity bench is not.
What would close it
Contributor history and at least two named engineers on the harness.
Enterprise-grade distribution and procurement
High
Team coverage today
Public policy pages exist, but trust artifacts and Google channel are not verified.
What would close it
Security packet, live Google listing, and buyer procurement references.
Consulting-specific workflow adoption
Critical
Team coverage today
Company narrative is clear, but retained usage is unknown.
What would close it
Paid retention cohorts and customer-controlled references.

Public Professional Footprint

These checks of public technical artifacts, professional social presence, and credential traces carry low decision weight by design; they sharpen founder-call questions but do not alone change the verdict.

Technical artifacts (repo level)
Aymeric Roucher
evidenced
Adrien Urlacher
unknown
What The Public Record Shows
Hugging Face smolagents and Aymeric Roucher GitHub evidence agent-layer depth; Folio-AI-Labs repositories show a thin public Folio footprint.
Professional social content
Aymeric Roucher
claimed
Adrien Urlacher
claimed
What The Public Record Shows
Aymeric Roucher LinkedIn and Adrien Urlacher LinkedIn support public association with Folio, but repeated founder claims are not validation.
Education / credentials
Aymeric Roucher
claimed
Adrien Urlacher
claimed
What The Public Record Shows
Public profiles include credential narratives; this review did not independently verify degree completion.
Publications / patents / certifications
Aymeric Roucher
unknown
Adrien Urlacher
unknown
What The Public Record Shows
Decision-bearing patents or certifications are not visible in the public record reviewed.

Net read: the public footprint supports a rare technical founder signal, but the visible Folio bench and founder commitment remain gating questions.

Founder-Market Fit Read

What is promising: Aymeric's agent-layer evidence is real and directly relevant to an AI-agent product.
What is missing: Independent proof that the team covers PowerPoint and Slides rendering fidelity, enterprise security, sales, and customer success.
What to prepare: Org chart, contributor history, founder time allocation, and independent technical validation of native-file editing.

VERSO.AI
Legal operating entity behind Folio.
French government registry
Evidencesource-backed fact
Confidencehigh
Follow-upProvide current registry extract and beneficial ownership details.
Verso.ai
Former or publisher-facing name visible in AppSource.
Microsoft AppSource Folio LinkedIn
Evidencesource-backed fact for publisher name; company claim for rebrand continuity
Confidencehigh
Follow-upReconcile contracting and marketplace names.
Folio-AI-Labs
Public GitHub organization for benchmark materials.
Folio-AI-Labs repositories
Evidencesource-backed fact
Confidencemedium
Follow-upExplain production repository ownership and engineering bench.
Aircall
Adrien Urlacher's concurrent public role context.
Adrien Urlacher LinkedIn
Confidencemedium
Follow-upConfirm whether Aircall role is current, ended, or part-time.

Traction

Customer Status Table

Air France-KLM demo material
Folio homepage uses public-material samples, not customer-controlled evidence.
No proofClaim onlyUnknownUnknownUnknownmedium
McKinsey public report sample
Folio homepage uses a public report sample, not customer-controlled evidence.
No proofClaim onlyUnknownUnknownUnknownmedium
Draghi report sample
Folio homepage uses a public report sample, not customer-controlled evidence.
No proofClaim onlyUnknownUnknownUnknownmedium

G2, Trustpilot, and Capterra were checked and showed no matching Folio customer reviews or case studies. These absences carry little weight because enterprise add-ins rarely surface on review platforms, and they do not prove zero paying customers.

Traction Signals

Microsoft AppSource listing with one public rating
source-backed facthigh
SourceMicrosoft AppSource
Installs, active organizations, conversion, and rating history.
Product Hunt product presence
source-backed fact for public page fields; company claim for maker commentshigh
SourceProduct Hunt
Paid conversion or user references.
LinkedIn company footprint
source-backed fact for public page fieldshigh
SourceFolio LinkedIn
Hiring, customers, or employee growth.
PrezEval public repository
source-backed fact for repository existence; company claim for benchmark validityhigh
SourcePrezEvalPublic GitHub
Independent rerun and held-out templates.

G2, Trustpilot, and Capterra were checked and showed no matching Folio listing on review platforms. These absences carry little weight for early-stage companies and do not prove zero traction.

Hiring And Org Momentum

LinkedIn company
source-backed factPublic company footprint shows very early scale.
SourceFolio LinkedIn
Consistent with a pre-seed, thin-bench company.

Indeed, Glassdoor, LinkedIn Jobs, and the Folio homepage were checked and showed no matching job postings or careers page. These absences carry little weight since job boards are often incomplete for pre-seed companies, and employee sentiment remains unavailable.

Role-mix read: The public record does not show a visible team build-out, so role mix is unknown.
Momentum read (growth / steady / contraction / unknown): unknown, with public signals consistent with a newly incorporated, pre-scale company.

Traction Quality Read

Traction Dimension	Current Status	Good Enough For First Call?	Needed For Deep Diligence
Product shipped	Confirmed for PowerPoint through Microsoft AppSource.	Yes.	Marketplace analytics and install-to-paid conversion.
Paid adoption	No independent public evidence found.	No, unless supplied before the call.	Billing cohorts, paid seats, references, and retention.
Retention	Not public.	No.	Cohort retention and renewal reasons.
Benchmark validity	Self-authored and not independently rerun.	No.	Blinded third-party rerun.
Distribution breadth	PowerPoint confirmed; Google Slides unverified.	Partial.	Live Google listing and OAuth verification.

Competitive Landscape

Segment	Examples	Customer Alternative	Pressure On Company
Incumbent suite AI	Microsoft Copilot, Google Gemini in Slides	Stay inside existing PowerPoint or Workspace subscription.	Bundled alternatives can compress price and absorb the workflow.
Direct add-in and native-file peers	Plus AI, Claude for Microsoft 365	Use another tool inside PowerPoint or Slides.	Contradicts category-of-one positioning and raises fidelity benchmark bar.
Adjacent AI presentation platforms	Gamma, Beautiful.ai, Presentations.ai, GenSpark, Decktopus	Generate, rewrite, or export slides outside Folio.	These vendors own search demand and user expectations.
Broad design and productivity substitutes	Canva, manual PowerPoint or Google Slides plus designers	Keep existing workflow or use a broader design suite.	Lowers willingness to pay for a standalone add-in unless editing fidelity is materially better.

Category Visibility Snapshot

Google / United StatesAI for PowerPoint

Captured page-one results (2026-06-15): Gamma, Microsoft, Slidesgo, Canva, and other category pages appeared in page-one results.
Was the company present?: No.

Google / United StatesAI slide generator

Captured page-one results (2026-06-15): Gamma, Slidesgo, Manus, Canva, Adobe Express, and GenSpark appeared in page-one results.
Was the company present?: No.

Google / United StatesAI presentation tool

Captured page-one results (2026-06-15): Gamma, Prezent, Canva, Slidesgo, Plus AI, and Adobe Express appeared in page-one results.
Was the company present?: No.

Google / United StatesAI for consulting decks

Captured page-one results (2026-06-15): Listicles and consultant-focused vendors such as Plus AI, GetAlai, SlideSpeak, and Gamma appeared in page-one results.
Was the company present?: No.

Google / FranceAI for PowerPoint

Captured page-one results (2026-06-15): Gamma, Manus, Presentations.ai, Slidesgo, Canva, and Microsoft appeared in page-one results.
Was the company present?: No.

Google / FranceAI slide generator

Captured page-one results (2026-06-15): Slidesgo, Manus, Canva, Microsoft, Gamma, and Adobe Express appeared in page-one results.
Was the company present?: No.

Google / FranceAI presentation tool

Captured page-one results (2026-06-15): Gamma, Presentations.ai, Canva, Microsoft, Adobe Express, and other tools appeared in page-one results.
Was the company present?: No.

Google / FranceAI for consulting decks

Captured page-one results (2026-06-15): Listicles and consultant-focused vendors such as Plus AI, Gamma, and Auxi appeared in page-one results.
Was the company present?: No.

Google Search in the United States and France was checked across AI presentation and consulting-deck category queries and Folio did not appear in page-one results. This absence carries medium weight for distribution because category search is dominated by larger rivals.

Visibility read (inference, low confidence): Folio had zero top-10 organic presence across the checked category searches, so near-term distribution likely depends on marketplaces, outbound, founder network, or direct consulting channels rather than category search.

Pricing And Competitive Benchmark

Alternative	What It Offers	Public Price Signal	Price Vs. This Company	Evidence Label
Folio	Credit-based AI slide editing and creation.	Folio pricing lists $0 to $499 monthly tiers.	Baseline.	`company claim`
Microsoft Copilot	Native Microsoft 365 AI in PowerPoint and broader suite.	Microsoft 365 Copilot pricing lists user-based pricing.	Bundle pressure because buyers may already buy Microsoft 365.	`company claim`
Google Workspace Gemini	Native AI inside Workspace tiers.	Google Workspace pricing lists Workspace plan pricing with Gemini availability.	Bundle pressure for Google organizations.	`company claim`
Gamma	AI presentation generation and editing workflow.	Gamma pricing lists paid plans.	Lower-cost presentation workflow alternative.	`company claim`
Beautiful.ai	Smart-slide design automation.	Beautiful.ai pricing lists Pro and Team plans.	Comparable specialist presentation software pricing.	`company claim`
Claude for PowerPoint	Anthropic add-in and Microsoft 365 workflow.	Claude help center states plan requirements.	Competes for the same in-PowerPoint work.	`source-backed fact`
Plus AI	PowerPoint and Google Slides add-in with native-file positioning.	Plus AI shows add-in positioning; detailed pricing was not central to this review.	Direct feature and buyer overlap.	`source-backed fact`

Price positioning read: Folio's credit model can be attractive if it replaces expensive professional editing time, but suite AI pricing and low-cost specialist tools create a tight willingness-to-pay band.
Price claims to correct or substantiate: Any claim that Folio is cheaper or materially better than Copilot, Gemini, Plus AI, or Gamma needs current third-party bake-off and total-cost evidence.

Competitive Wedge

Folio's possible wedge is faithful editing of existing native decks for consulting teams, not generic slide generation. That wedge could be defensible if it combines technical fidelity, team-specific templates, brand rules, stored instructions, and review workflows that make renewal stronger over time. Today, that wedge is mostly company claim: direct and adjacent competitors are visible, Folio has no category search presence, and the benchmark advantage has not been independently validated.

Risks And Open Questions

Risk	Severity	Evidence	What To Ask
No independent paying-retention evidence	Critical	Public checks found no paying customer, revenue, retention, or active-org proof.	Show billing cohorts, references, and net revenue retention.
Platform-owner absorption	Critical	Microsoft Marketplace Terms, Google Workspace Marketplace Developer Agreement, Microsoft Copilot, and Google Gemini in Slides show dependence and substitutes.	What value remains when native agents improve?
Self-authored benchmark	High	Folio benchmark blog and PrezEvalPublic GitHub.	Can an independent party rerun it blinded?
Thin visible team	High	Folio LinkedIn, Folio-AI-Labs repositories, and Adrien Urlacher LinkedIn.	Who builds and sells full-time?
EU procurement readiness	Medium-high	Folio privacy policy and Folio pricing.	Can EU consulting buyers procure today?
Market ceiling	High	BLS, empower PowerPoint study, and Folio pricing.	What proves expansion beyond consulting?

Pre-Mortem: The Most Likely Obituary

Cause of death (one sentence): Folio died because Microsoft and Google shipped good-enough native deck-editing agents, and Folio never converted its self-authored fidelity edge into paid retention or durable switching costs.
The causal chain (3-5 steps from today to the shutdown): First, the company failed to produce a logo-independent paying-retention base. Second, the thin bench slowed progress on messy enterprise-template fidelity. Third, a bundled native agent removed procurement friction and became adequate for most users. Fourth, Folio's stored instructions and template memory did not become deep switching costs. Fifth, the team could not fund a long enough fight from a very early-capitalized entity.
The earliest observable warning sign: The next founder conversation still produces benchmark demos and future-interest language, but no paid cohort and no referenceable consulting team.
The question that defuses this chain today: Show one consulting team that has paid and renewed, and explain exactly what that team would lose by switching to a bundled native agent next quarter.

Decision-Critical Unknowns

Unknown	Why It Is Decision-Critical	Best Evidence	Decision Effect
Paying customers and retention	It determines whether adoption is real.	Billing cohorts, paid seats, references, and renewal data.	Real retention can move toward pursue; zero paid usage after months live moves to pass.
Durable value capture	It determines whether the company keeps value against suite owners.	Win-loss, pricing power, switching costs, and renewal reasons after native-agent competition.	A credible answer preserves the option; no answer moves to pass.
Benchmark validity	It determines whether differentiation is real.	Blinded third-party rerun on held-out enterprise templates.	A material win strengthens the thesis; parity or worse weakens it sharply.
Team coverage	It determines whether the hardest build can survive beyond one founder.	Org chart, contributor history, and founder time commitments.	Bench depth can move toward pursue; sole-engineer dependency can move to pass.
Google channel	It determines whether cross-platform distribution is real.	Live Workspace Marketplace listing and verification status.	Confirmed listing helps distribution; no listing weakens trust.
Cap table and runway	It determines whether a first check can buy a defendable option.	Cap table, cash, burn, and prior instruments.	Clean entry helps; no runway or messy instruments reduce conviction.

Diligence Questions

First Call

Question	Why It Matters	Good Evidence
How many paying accounts, seats, MRR or ARR, and retained cohorts exist today?	It is the central adoption gate.	Billing exports, cohort retention, and reference permissions.
What does Folio durably capture that Microsoft or Google will not absorb?	It is the value-capture kill question.	Win-loss against native agents, pricing power, and switching-cost evidence.
Can PrezEval be rerun blinded on held-out enterprise templates?	Differentiation is self-authored today.	Independent rerun protocol, raw outputs, and reviewer access.
Who builds the slide harness besides Aymeric, and is Adrien full-time?	The hardest build and GTM coverage are not fully visible.	Org chart, contributor history, founder schedule, and equity split.
Where is the live Google Workspace listing and what are marketplace active-org counts?	Distribution claims need reality checks.	Publisher dashboards and live marketplace URL.

Follow-Up

Question	Why It Matters	Good Evidence
What are LLM costs, provider terms, and gross margin by plan?	Model dependence can compress margin.	COGS bridge, provider contracts, and BYOK usage.
Can a regulated EU buyer procure today?	Consulting buyers may require trust artifacts.	DPA, subprocessor terms, EU hosting plan, and attestation roadmap.
What evidence shows expansion beyond the consulting wedge?	The initial modeled SAM may be too small.	Adjacent-segment retained cohorts or paid pilots.
What is the cap table and current runway?	Ownership and follow-on math depend on it.	Cap table, cash, burn, and financing plan.
How will brand and entity names reconcile for enterprise contracting?	Folio, Verso.ai, and VERSO.AI appear across surfaces.	Registry extract, publisher migration plan, and contracting statement.

Kill Criteria

Kill Criterion	Evidence That Would Trigger It
Confirmed zero paying customers after months live.	Founder materials show no paid accounts, no retained cohorts, and only trials or curiosity usage.
No credible value-capture answer versus bundled native agents.	Founder answer rests on "better product wins" without win-loss, pricing power, or switching-cost evidence.
Blinded benchmark shows parity or worse.	Independent rerun fails to show a material editing-fidelity advantage.
Hardest build is confirmed sole-engineer-dependent with no hire plan.	Org chart shows Aymeric is the only harness owner and Adrien is not full-time.
Google channel is not live and AppSource active-org counts are negligible.	Marketplace evidence shows distribution is mostly theoretical.
EU consulting buyers cannot procure and there is no remediation path.	No DPA, no residency option, no live security controls, and no attestation plan.

Double-Down Criteria

Double-Down Criterion	Evidence That Would Justify More Diligence
Paying retention is real.	Billing cohorts show retained paying teams and referenceable consulting users.
Editing fidelity is independently validated.	A blinded rerun shows Folio materially outperforms native agents on held-out enterprise templates.
Capture survives the bundle.	Customers renew despite Copilot or Gemini availability because of accumulated Folio assets.
Team and runway are fundable.	At least two harness engineers, full-time founders, clean cap table, and runway to the next milestone.
Market expands beyond consulting.	Adjacent deck-heavy roles show retained usage or paid conversion.

Founder Action Plan

Timeframe	Action	Output
Before the next investor call	Assemble an independent paying-adoption packet.	Billing cohort, paid seats, net revenue retention, and two or three references.
Before the next investor call	Set up a blinded third-party PrezEval rerun.	Held-out templates, protocol, raw outputs, and reviewer access.
Before the next investor call	Write a value-capture brief.	Win-loss versus Copilot and Gemini, pricing-power evidence, and switching-cost explanation.
Before the next investor call	Publish the team coverage map privately to investors.	Org chart, contributor history, founder time allocation, and hiring plan.
Before deeper diligence	Provide marketplace analytics and Google listing proof.	AppSource installs and active organizations, conversion data, and live Workspace Marketplace URL.
Before deeper diligence	Complete a procurement packet.	DPA, subprocessor list, EU data plan, security controls, and attestation timeline.
Before term-sheet discussion	Provide financing materials.	Cap table, current cash, burn, prior instruments, and milestone budget.

Decision

Screen: hold.
Confidence: medium.
Rationale: Folio earns a time-boxed call because one founder's agent-framework authorship is independently corroborated and the editing-fidelity wedge is plausibly non-obvious, but it is not yet underwritable because adoption, retention, benchmark validity, team coverage, Google distribution, and value capture are unresolved.
What would move this to pursue: Independent paying retention, a blinded benchmark win, a credible capture answer versus suite-native agents, at least two evidenced harness engineers, full-time founders, clean cap table, and expansion evidence beyond the consulting wedge.
What would move this to pass: Confirmed zero paid retention after months live, no durable capture answer, benchmark parity, sole-engineer dependency, no Google channel with negligible AppSource penetration, or no founder proof packet by 2026-09-15.
Recommended next step: One founder call, gated on a pre-supplied proof packet, with a pass-by date of 2026-09-15.
Founder preparation standard: Treat every public claim as a diligence item and bring proof before narrative.

How We Would Miss This One

The miss scenario: Folio could be the rare early company where a genuinely exceptional agent builder solves the tedious native-deck editing layer before incumbents care, and the fund over-weights today's absence of public traction because the company is only months old.
Flip conditions: The miss case becomes real if Folio shows retained paying consulting teams, a blinded fidelity win, switching costs from accumulated per-account assets, full-time team coverage, and adjacent-segment pull.
Revisit trigger / date: Revisit by 2026-09-15 or sooner if the founders provide the proof packet covering paid retention, benchmark validation, team coverage, marketplace analytics, and value capture.

Source Log

Folio homepage
get-folio.ai
Product positioning, consulting wedge, benchmark claims, BYOK and enterprise claims, demo samples.
Retrieved 2026-06-15company-controlled pagehigh
Folio pricing
get-folio.ai
Credit tiers, $0 to $499 monthly pricing, overage rates, and single sign-on plan gating.
Retrieved 2026-06-15company-controlled pagehigh
Folio help center
get-folio.ai
Product workflow and Google Slides claim needing validation.
Retrieved 2026-06-15company-controlled pagemedium
Folio terms
get-folio.ai
Legal baseline, governing law, liability, and service terms.
Retrieved 2026-06-15company-controlled legal pagehigh
Folio privacy policy
get-folio.ai
Sub-processors, data-processing posture, LLM provider dependence, and EU procurement questions.
Retrieved 2026-06-15company-controlled legal pagehigh
Folio LinkedIn company page
linkedin.com
Public company footprint, follower and employee-range context, and brand history.
Retrieved 2026-06-15LinkedIn company profilehigh
Product Hunt Folio AI
producthunt.com
Public launch/community signal and cofounder association.
Retrieved 2026-06-15community platformhigh
Microsoft AppSource Folio AI
appsource.microsoft.com
PowerPoint listing, publisher name, public rating, and distribution confirmation.
Retrieved 2026-06-15platform marketplace listinghigh
Microsoft Office add-in install page
pages.store.office.com
AppSource asset identifier and Verso.ai publisher evidence.
Retrieved 2026-06-15platform install pagehigh
French government registry VERSO.AI
annuaire-entreprises.data.gouv.fr
Incorporation date, legal entity, capital, Paris address, and non-employer status.
Retrieved 2026-06-15official registryhigh
Aymeric Roucher LinkedIn
fr.linkedin.com
Founder identity and self-reported background.
Retrieved 2026-06-15LinkedIn person profilemedium
Adrien Urlacher LinkedIn
linkedin.com
Founder identity, Folio association, and concurrent Aircall role signal.
Retrieved 2026-06-15LinkedIn person profilemedium
Aymeric Roucher GitHub
github.com
Public engineering footprint and Hugging Face organization affiliation.
Retrieved 2026-06-15GitHub profilehigh
Hugging Face smolagents
github.com
Independent corroboration of Aymeric Roucher's agent-framework authorship.
Retrieved 2026-06-15GitHub repositoryhigh
PrezEvalPublic GitHub
github.com
Public benchmark materials and validation pathway.
Retrieved 2026-06-15GitHub repositoryhigh
Folio-AI-Labs repositories
github.com
Public repository footprint and engineering-bench question.
Retrieved 2026-06-15GitHub organization pagemedium
Folio benchmark blog
get-folio.ai
Company-authored benchmark claims.
Retrieved 2026-06-15company-controlled blogmedium
Folio PowerPoint engineering blog
get-folio.ai
SimpleXML and native-editing claims.
Retrieved 2026-06-15company-controlled blogmedium
Folio connector blog
get-folio.ai
Connector-layer moat narrative.
Retrieved 2026-06-15company-controlled blogmedium
Folio Copilot comparison
get-folio.ai
Company-authored Copilot comparison and pricing-copy inconsistency.
Retrieved 2026-06-15company-controlled blogmedium
Folio best AI tools blog
get-folio.ai
Company-authored competitor positioning and benchmark narrative.
Retrieved 2026-06-15company-controlled bloglow
Microsoft Copilot
microsoft.com
Incumbent AI productivity competitor and bundle context.
Retrieved 2026-06-15incumbent company pagehigh
Microsoft AI PowerPoint generator
microsoft.com
PowerPoint-native AI workflow context.
Retrieved 2026-06-15incumbent company pagehigh
Microsoft 365 Copilot pricing
microsoft.com
Bundle pricing pressure.
Retrieved 2026-06-15incumbent company pricing pagemedium
Microsoft Marketplace Terms
learn.microsoft.com
Marketplace discretion, throttling, and channel risk.
Retrieved 2026-06-15platform termshigh
Microsoft annual report 2025
microsoft.com
Microsoft suite scale, segment revenue, and Copilot context.
Retrieved 2026-06-15public-company filinghigh
Google Slides product page
workspace.google.com
Google Slides surface context.
Retrieved 2026-06-15incumbent company pagehigh
Google Gemini in Slides
workspace.google.com
Native Google slide AI and incumbent substitute context.
Retrieved 2026-06-15incumbent company pagehigh
Google Workspace pricing
workspace.google.com
Workspace bundle pricing and Gemini tier context.
Retrieved 2026-06-15incumbent company pricing pagemedium
Google Workspace Marketplace Developer Agreement
workspace.google.com
Google marketplace removal and channel-risk terms.
Retrieved 2026-06-15platform termshigh
Claude for Microsoft 365
claude.com
Direct/adjacent PowerPoint workflow competitor.
Retrieved 2026-06-15competitor company pagehigh
Claude help center PowerPoint
support.claude.com
Plan requirements and PowerPoint feature context.
Retrieved 2026-06-15competitor help centerhigh
Plus AI
plusai.com
Direct add-in competitor and consulting/professional slide workflow overlap.
Retrieved 2026-06-15competitor company pagehigh
Crunchbase Plus Docs
crunchbase.com
Plus AI company scale and funding context.
Retrieved 2026-06-15startup database (Crunchbase)medium
GenSpark
genspark.ai
Adjacent AI workspace and PowerPoint plugin competitor.
Retrieved 2026-06-15competitor company pagehigh
Crunchbase GenSpark
crunchbase.com
GenSpark funding and scale context.
Retrieved 2026-06-15startup database (Crunchbase)medium
Gamma
gamma.app
Adjacent AI presentation competitor and consultants positioning.
Retrieved 2026-06-15competitor company pagehigh
Gamma pricing
gamma.app
Specialist presentation-tool pricing benchmark.
Retrieved 2026-06-15competitor company pricing pagemedium
Crunchbase Gamma
crunchbase.com
Gamma scale and investor context.
Retrieved 2026-06-15startup database (Crunchbase)medium
Beautiful.ai
beautiful.ai
Adjacent presentation competitor.
Retrieved 2026-06-15competitor company pagehigh
Beautiful.ai pricing
beautiful.ai
Specialist presentation-tool pricing benchmark.
Retrieved 2026-06-15competitor company pricing pagemedium
Crunchbase Beautiful.ai
crunchbase.com
Beautiful.ai funding and scale context.
Retrieved 2026-06-15startup database (Crunchbase)medium
Presentations.ai
presentations.ai
Adjacent AI presentation competitor.
Retrieved 2026-06-15competitor company pagemedium
Crunchbase Presentations.ai
crunchbase.com
Presentations.ai investor context.
Retrieved 2026-06-15startup database (Crunchbase)medium
Decktopus
decktopus.com
Adjacent AI presentation competitor.
Retrieved 2026-06-15competitor company pagemedium
Crunchbase Decktopus
crunchbase.com
Decktopus company scale context.
Retrieved 2026-06-15startup database (Crunchbase)medium
Canva
canva.com
Broad design and presentation substitute.
Retrieved 2026-06-15competitor company pagehigh
Canva pricing
canva.com
Broad design-suite pricing benchmark.
Retrieved 2026-06-15competitor company pricing pagemedium
Crunchbase Canva
crunchbase.com
Canva scale context.
Retrieved 2026-06-15startup database (Crunchbase)medium
Crunchbase Magical Tome
crunchbase.com
Closed-status category churn signal.
Retrieved 2026-06-15startup database (Crunchbase)high
BLS Management Analysts
bls.gov
Bottom-up consulting/strategy seat denominator.
Retrieved 2026-06-15government statisticshigh
FRED management consulting revenue
fred.stlouisfed.org
U.S. strategic management consulting revenue context.
Retrieved 2026-06-15government statistics via FREDhigh
Coherent Market Insights presentation software market
coherentmarketinsights.com
Broad presentation-software TAM ceiling.
Retrieved 2026-06-15market-sizing aggregatorlow
Research and Markets global presentation software market
researchandmarkets.com
Second broad presentation-software TAM ceiling.
Retrieved 2026-06-15market-sizing aggregatorlow
empower PowerPoint study
empowersuite.com
Slide-usage and formatting-pain inputs for bottom-up model.
Retrieved 2026-06-15vendor surveylow
Goldman Sachs AI agents
goldmansachs.com
Agentic-software timing and value-capture context.
Retrieved 2026-06-15named analyst reportmedium
Gartner agentic AI forecast
gartner.com
Agentic AI adoption and cancellation-risk timing.
Retrieved 2026-06-15named analyst reportmedium

Access limitations: Public sources did not expose Folio's paid-seat count, MRR or ARR, retention, AppSource installs, active organizations, Google Workspace listing, cap table, cash, burn, customer references, LLM cost structure, or executed security documents. Search-result snapshots and AI-answer visibility checks were used only as low-weight visibility or absence signals and are not listed as citable URLs. Some databases and company profiles may require login for full detail, so funding, employee, traffic, and investor context from those sources should be treated as directional unless confirmed directly with the company.

Public disclaimer: This is an independent public teardown based solely on publicly available information retrieved on 2026-06-15. It is research support, not investment, legal, tax, or financial advice. Company-provided statements are labeled as claims, not verified facts. Folio did not participate in or review this report. Errors and omissions are possible; final decisions remain with the reader.

What The Company Does#

Key Takeaways#

What Makes This Potentially Fundable#

The Four Holes To Close Before Fundraising#

Decision Snapshot#

IC Disagreement Map#

Scenario Range#

What Investors Will Test#

Claim Reconciliation: Inconsistencies Investors Will Catch#

Diligence Findings: Issues To Fix Before You Raise#

Data Room Readiness Checklist#

First-Call Agenda For The Startup#

Business Model And Revenue Signals#

Revenue Quality Checklist#

Funding And Ownership Context#

Founder, Team, And Related Entities#

Founder And Team#

Founder Competency Coverage#

Public Professional Footprint#

Founder-Market Fit Read#

Related Entities And Founder-Associated Companies#

Traction#

Customer Status Table#

Traction Signals#

Hiring And Org Momentum#

Traction Quality Read#

Competitive Landscape#

Category Visibility Snapshot#

Pricing And Competitive Benchmark#

Competitive Wedge#

Risks And Open Questions#

Pre-Mortem: The Most Likely Obituary#

Decision-Critical Unknowns#

Diligence Questions#

First Call#

Follow-Up#

Kill Criteria#

Double-Down Criteria#

Founder Action Plan#

Decision#

How We Would Miss This One#

Source Log#