Four Fates of Professional Workflow SaaS in the Age of AI

A Deep Comparison of Paradigm Shifts for INTU / ADBE / ADSK / PTC

Analysis Date: 2026-04-23

Navigation and How to Read This Report

Framework Dictionary Upfront

This report is not written according to traditional financial templates; instead, it analyzes companies based on a "business control points" framework. The complete dictionary is presented first, before moving into the main body of the report.

What Exactly Are I1-I8?

I1-I8 are not financial line items, but rather 8 "business control points."
You can think of them as: which layer a company truly controls in the age of AI.

Why Are I4 / I7 / I8 Particularly Important?

This report most frequently refers to I4 / I7 / I8, as these three items best distinguish between a company that is "just a tool" and one that "truly controls the workflow."

Therefore, when you see a company with strong I4 + I7 + I8, do not interpret it as merely having a "high framework score," but rather understand it as:

This company not only controls the entry point, but also bears responsibility, and furthermore, must step in to handle things when errors occur.
Such companies are typically harder to bypass in the age of AI.

What Are P1 / P4 / P11?

These three are not scores, but rather three different company "archetypes."

What is AI Asymmetry / Sequencing / Stack Coherence?

These three terms are not separate business functions but supplementary assessments of "transformation quality."

If You Only Want a Summary Table First

The sole purpose of this table is to first differentiate these four companies from being categorized as "the same type of software stock."
All subsequent chapters essentially answer the same question:

At which layer of the software workflow are these four companies positioned, and is AI helping them deepen this layer, or is it circumventing it?

Core Judgement

The market continues to price INTU / ADBE / ADSK / PTC as "a family of high-profit, vertical SaaS companies under AI pressure," but the latest paradigm shift framework reveals that they actually occupy four completely different control point tiers:

• INTU → Authority + Distribution Owner, which is an SME financial operating system with "both responsibility assumption and default distribution."
• PTC → Operational Control Layer, which is a product lifecycle execution layer that "truly controls processes and governance."
• ADSK → Workflow Default is moving towards the Control Layer, meaning "the default entry point is still in its hands, but the governance layer is still maturing."
• ADBE → Legacy Workflow Platform Defending with AI, meaning "the legacy platform relies on AI to retain core users, but entry points have partially shifted outward."

Pricing these four companies using the same P/E language essentially forces four different assets into a single template. The task of this report is to analyze these four assets separately and then re-sort them.

Five Core Views

Glossary

Chapter 1: Applying This Framework to Four Companies: A Visual Introduction

1.1 What Three Questions This Chapter Addresses First

1.2 Why the Market Puts These Four Companies in One Bucket

A market snapshot of the four companies on 2026-04-23 looks like this:

Surface similarities are extremely high:

The problem with this template isn't that it's wrong, but that it stops at financial commonalities and fails to capture differentiation in business control points. The impact of AI, for instance, affects Photoshop / QuickBooks / Revit / Windchill in entirely different ways; likewise, for high-gross-margin software, there are vast differences in who truly owns the I2 decision context / I3 execution / I4 authority / I8 entry point.

1.3 How This Framework Uncovers the True Structure of the Four Companies

The paradigm shift framework doesn't ask "does this company have AI" or "is this company impacted by AI"; it asks:

Intuition Trap: Simplifying AI enhancement and AI impact into unidirectional assumptions is the most common mistake. AI's true impact on a company depends on the interplay of the answers to these five questions:

• If a company's I2 has accumulated historical data across entities and cycles, AI training will strengthen it further (INTU's GenOS is based on 860 million filings).
• If a company's I4 bears compliance or audit responsibility, an LLM cannot directly replace it (IRS e-filer identity cannot be given to ChatGPT).
• If a company's I8 is at the discovery layer, it is most vulnerable to being co-opted by general LLMs / new tools (the logic of Canva acquiring Express).
• If a company genuinely triggers state changes in I3 (BOM modification / tax submission / construction drawing approval), tools may be replaceable, but the workflow continuity is hard to replace.

This is why it's essential to dissect by control points rather than label by AI tags.

1.4 Framework Introduction: Visualizing 8 Invariants

The starting point of the paradigm shift framework is to dissect a company's value creation into 8 invariants (I1-I8). These are not financial metrics, but rather dimensions of business control points:

Understanding Each Invariant in One Sentence:

1.5 Archetype Library: Four Structures Corresponding to Four Companies

In the v1.0 Archetype Library, four companies correspond to fourdifferent archetypes:

Key: Although ADSK and ADBE are both classified as P11, their positions differ — ADSK is a P11 → P1 migration path sample, ADBE is a pure P11 defensive sample. Chapters 3 and 4 of this report will dissect this difference: Their I4 / I8 / Stack Coherence have a structural gap of about 5 points and cannot be covered by the same label.

1.6 BSM: Why "Aggregate Scoring" Can Deceive You

BSM (Business Split Mechanism) is a mandatory discipline of the framework: any business line with revenue >5% of a company's total must be analyzed separately.

BSM Fate of the Four Companies:

Not performing BSM scoring is equivalent to saying "INTU as a whole = GBS's moat + CK's growth + Mailchimp's drag" averaged into a meaningless median.

1.7 Four Gates + Modifiers: Scoring Does Not Equal Conclusion

The framework requires each company to pass through four gates:

Scoring is not the conclusion; the mechanism is. This report requires a clear explanation for every score of every company: where the score comes from, why it isn't higher, and why it isn't lower.

Chapter 2: INTU — Not Tax Software, But an SMB Financial Operating System with Authority

Core Judgment: INTU's true identity is not a "vertical SaaS leader," but an **American SMB financial operating system that simultaneously possesses I4 authority, I7 exception and I8 distribution**. The combined scores of the other three companies in these three categories do not surpass INTU's individual score. This is the fundamental difference between P4 and P11, and the only true reason INTU deserves to be revalued.

2.1 Conclusions First

2.2 Five-Layer Business Structure

While INTU will merge its segments into a single "Consumer" disclosure starting in FY2026, it structurally remains a five-layer business:

2.3 I1 — What Customers Are Really Paying For

Compared to the other three: ADBE/ADSK/PTC only have the Access layer (I1 = 8/11 soft cap triggered). Only INTU simultaneously has:

• Outcome — CK earns commissions through loan leads; revenue is tied to whether the client **obtains a loan**.
• Authority Premium — TurboTax Max Defend ($80) / Error Insurance / Compliance premium from IRS authorized e-filer status.
• Distribution Premium — 46,000 CPA firms recommend QuickBooks at a 75% rate; this is not subscription revenue, but a distribution structure.

This is why INTU's I1 is 9.5/11, while the other three top out at 8/11: It transcends the soft cap of pure access.

2.4 I2 — Deepest Financial Context

Why this category scores 8.5/9: No competitor simultaneously possesses a multi-year history across all four dimensions: **Tax + Accounting + Credit + Accountants**. Xero has accounting but no tax, H&R Block has tax but no year-round credit, NerdWallet has credit but no transaction flow. INTU's I2 is a true **cross-entity, cross-cycle closed loop**.

2.5 I4 + I7 — The True Chasm Between INTU and the Other Three

This subsection is one of the most important parts of the entire report, because it explains why INTU is P4 and the other three are not.

Quantification Comparison:

Conclusion: INTU's I4+I7 total is 42% higher than the second-ranked PTC. This is not a quantitative difference; it's a qualitative difference. INTU is truly **assuming responsibility**, while the other three are **providing tools**.

Why this difference is valuable:

1. In the AI era, tools will be commoditized, but **assuming responsibility** will be at a premium. LLMs can achieve 90% accuracy in PDF search, but an LLM cannot become an IRS e-filer.
2. The migration cost for tools is "learning a new interface." The migration cost for responsibility is "re-obtaining regulatory authorization + rebuilding distribution networks + re-purchasing actuarial insurance data."
3. The market's current premium for INTU (18x P/E) vs. ADBE (9.6x P/E) primarily prices in the I1 growth differential. It **has not yet fully priced in the structural difference of I4**.

2.6 I8 — Dual-Layer Entry Points of the Distribution Network

Quantification:

• TurboTax 62% federal e-file market share
• QuickBooks 62-80% U.S. SMB market share
• Credit Karma: 37M MAU
• 46,000 CPA firms, 75% recommend QuickBooks

Comparison: ADBE has creative entry points but discovery has shifted externally. ADSK has AECO entry points but the discovery layer is under AI threat. PTC has domain-specific multiple entry points but a small scope. Only INTU has default entry points simultaneously in both B2C (year-round + tax season) and B2B (SMB + accountants).

2.7 Why AI acts more as an accelerator than a substitute for INTU

Key Judgment: AI cannot directly replace TurboTax, because TurboTax's core value is not "tax calculation capability" (LLMs are already strong enough), but rather IRS authorization + audit defense + error insurance + historical migration costs. AI places 50% weight on the dissolvent side, but 50% weight on the enhancer side (especially TT Live human replacement), resulting in a net score of +4 (Neutral Positive).

2.8 INTU Main Table

2.9 INTU Kill Switch

Chapter 3: ADBE — A Standard Sample of P11, and a Textbook Example of "AI Split Personality"

Core Judgment: ADBE is neither dying nor winning. It is **six completely different AI destinies** being compressed by the market into a single AIAS score. Separating these six lines reveals: **Creative Cloud Consumer is bleeding, but Document Cloud and GenStudio are gaining market share; Firefly's strategy has shifted from "building the strongest model" to "creating the best model supermarket."**

3.1 Key Takeaways of this Chapter

3.2 AI Split Personality — Six Business Line Heatmap

Key Data Points:

• CC Pro: 62% of users call Firefly "essential"
• CC Consumer: Canva 265M MAU >> Express 80M MAU
• Firefly: ARR >$250M, QoQ +75%
• Document Cloud: Business Pro +16% YoY
• GenStudio: ARR >$1B, +30%, Top 50 clients 90% adopt AI-first

Structural Observation: Averaging the 22 AIAS scores (from -9 to +13) into an "overall company AIAS +0.42 red team adjusted" essentially **erases information**. The true diagnostic method is to **weight each business line and examine its AI destiny**, then assess whether the revenue structure shift can occur before the declining lines are exhausted — this is the correct way to approach the P11 split personality.

3.3 The Divergent AI Destinies of CC Pro vs. CC Consumer

This chart explains the true meaning of the P11 Defensive Growth Variant — the key is to differentiate "what to defend / what not to defend / what to attack":

• Defend: CC Pro professional creation + Document Cloud enterprise PDF + GenStudio content supply chain
• Do not defend / Concede: CC Consumer / Express → Let Canva take over; Adobe is **strategically retreating** from the individual creator market
• Attack: Firefly model supermarket + Experience Cloud agentic automation

3.4 I1-I8 Master Table + Mechanisms

3.5 Strategic Significance of Firefly's "Model Supermarket" Strategy

Core Insight: If Midjourney releases a better model tomorrow, Adobe only needs to integrate it in the next update; models are commodities, workflows are infrastructure. This is Adobe's true defense in I3 + I8 – not contending "whose model is best," but contending "who owns the default starting and ending points of professional creative workflows."

3.6 Why ADBE's I8 is the Weakest Among the Four Companies

Quantification: Canva 265M MAU vs Express 80M MAU; Midjourney is estimated to have 20M+ active users, Runway several million. Adobe has lost significant traffic at the "first click of work" step. This is the core reason why I8 dropped from ADSK's 4/5 to ADBE's 2.5/5.

3.7 ADBE Master Table + Differentiating Points

Structural Diagnosis: ADBE is flat in I6 (economic engine equally strong), but is entirely weaker than the other three companies in I4 / I7 / I8. This is a typical symptom of P11 — the economic engine steps down before the control points, with franchise quality supporting the P/E.

3.8 ADBE Kill Switch

Chapter 4: ADSK — The King of Workflow Default, Moving Up the Control Layer

Core Judgment: Both ADSK and ADBE are categorized as P11, but they are not in the same P11 position. ADBE is a pure defensive P11 sample (discovery has shifted outward), while ADSK is a sample on the P11→P1 migration path (discovery is still retained, and authority is slowly emerging). This structural gap of approximately 5 points, if covered by a single "P11 defensive growth" label, would lead to misaligned investment judgments. This chapter will clarify the evidence and mechanisms of this gap.

4.1 This Chapter's Conclusions First

4.2 Design & Make Platform Four Quadrants

Key Data:

• FY2027 guidance revenue +12-13%, FY26 reported +18% including billing transition catch-up
• 97% of revenue is recurring
• Non-GAAP OPM 38%, FCF Margin 33.4%
• Billings +30% shows active new subscriptions but with transition noise

4.3 BIM Mandate — Umbrella or Misunderstood?

Key Insight: The mandate is derived protection, not direct protection. The government mandates BIM workflows, not Revit products. This makes ADSK's I5 one notch stronger than ADBE's but weaker than PTC's Codebeamer FDA-21-CFR-Part-11 / AS9100 compliance stack — because PTC handles actual tooling-for-compliance, while ADSK handles only tooling-for-a-mandate-that-requires-BIM. This distinction between derived vs. direct is key to understanding why the BIM mandate is only a high-probability moat, not a definitive one.

4.4 I1-I8 Master Table + Mechanisms

4.5 P11→P1 Migration Path

Key Evidence:

• APS updated on 2025-12-07, providing data APIs + MCP servers + marketplace + usage model for agentic AI
• Revit Assistant Tech Preview launched on 2026-04-22, capable of querying models / generating schedules / automating documentation

4.6 Why ADSK is "More Stable in the Same Tier" than ADBE

4.7 ADSK Kill Switch

Chapter 5: PTC — The Industrial Software Most Resembling a "True Control Layer" Among the Four Companies

Core Assessment: PTC is not a patchwork of industrial software, not merely a CAD company, nor solely a PLM subscription provider. Post the divestment of Kepware/ThingWorx, it is converging PLM + ALM + SLM + cloud CAD into an Intelligent Product Lifecycle Layer. This is the only company among the four where the primary archetype is P1 rather than P11. Owner Economics are the cleanest (SBC 4.8%), and I4 has already incorporated the governance stack of TÜV Trusted Tool and FDA 21 CFR Part 11.

5.1 Key Takeaways for This Chapter

5.2 Intelligent Product Lifecycle — A Unified Control Layer Across Design/Change/Service

Key Data:

• FY2025 Revenue $2.74B, ARR $2.479B, recurring $2.601B
• Non-GAAP OPM 48% (Highest among the four)
• SBC/Revenue 4.8% (Lowest among the four, half of ADSK's 10.9%)
• FCF $857M, FY26 Guidance $850M (excl. divested)
• 2026-03-16 Kepware+ThingWorx Divestment completed, Net Debt/EBITDA Decreases from 2.28x to 1.05x

5.3 BSM Structure Comparison Before and After IoT Divestment

Key Observation: The true value of the IoT divestment is not a "0.5pp increase in growth rate," but rather the **purification of the core structure** — the PLM family becomes the 62% core, with P1 attributes overshadowing P11. This is why one cannot solely judge divestment value by the ARR difference before and after (only 50bp); the real change is a **shift in archetype focus**, which has a far greater impact on valuation language selection than 50bp ARR.

5.4 Windchill — Mechanisms of Governed Execution

Why this is governed execution, not just a tool:

1. Changes are not suggestions, but actual state modifications: BOM / part master / ECO in Windchill is the single source of truth
2. Embedded Audit Stack: TÜV certified Codebeamer as a Trusted Tool, meeting IEC 62304 / ISO 26262 / FDA 21 CFR Part 11
3. BMW Case Study (2026-04-01): BMW Group's enterprise-level standardization of Codebeamer illustrates that "governed requirements engineering" has been standardized within large clients.

5.5 I4 — Where PTC Excels Over ADSK

Quantification:

5.6 I6 Owner Economics — PTC is the Cleanest Among the Four

Structural Implications: PTC's Non-GAAP OPM of 48% is almost exactly the true Owner OPM. The Non-GAAP OPMs of the other three companies need to be discounted for SBC. This is why PTC's FWD PE of 19.4x is not considered cheap — its "Standard PE" is essentially its "Owner PE".

5.7 PTC Master Table

5.8 PTC Kill Switch

Chapter 6: Cross-Company Comparison of Four Companies — Placing the 8 Invariants on the Same Table

6.1 How to Read This Chapter Most Quickly

If you have already read the previous four chapters, the purpose of this chapter is to bring all judgments back to the same table.
If you don't have time to read the full text, then focus on these three tables:

1. 6.2 — Overall Table of 8 Invariants for Four Companies
2. 6.3 — Who Occupies Which Layer of Control Points
3. 6.6 / 6.7 — Budget Quality, Validation Quality, and Expansion Quality

Together, these three tables essentially form the "compressed conclusion" of the entire report.

6.2 Comparison Table of 8 Invariants

6.3 Layering of Control Points — Who Owns Each Layer

Key Observations:

1. S1 Discovery: Collective attrition to LLM — All four are under pressure, with ADBE being the most severe.
2. S2 Initiation — INTU/ADSK/PTC maintained, ADBE partial attrition (Canva/Figma).
3. S3 Execution — All four are still maintained, because real state changes still require professional tools.
4. S4 Completion — All four are maintained, this is the fundamental basis of P11 defense.
5. S5 Post-action — Only INTU and PTC have substantive coverage, ADBE/ADSK are largely absent.

6.4 AI Asymmetry Spectrum

Key Insight: Treating ADBE as a single score would overestimate the true risk. The worst of its six business lines is -9, the best is +13, with a weighted average of +0.42, which is the **split entity's pivot**; the investment implication depends on whether the absolute scale of GenStudio / DC / Firefly can shift revenue mix before CC Consumer collapses.

6.5 Sequencing Gate Comparison

Why INTU is Good:

• I1 vs I6: Outcome + Authority premium has been truly monetized in CK/TT Live, not just a narrative.
• I8 vs I5: Distribution (CPA network) has brought in B2B budget, not a mere story.
• I3 vs I4: Authority and Execution are strengthening synchronously.

Why the other three are Neutral and not Good: All have a "craftsmanship stronger than pricing" structure (high I6), but **pricing has not truly monetized a new level**. ADBE has a GenStudio prototype but only $1B, ADSK has APS/Flex but it remains marginal, PTC has Onshape+Codebeamer but its CAD leg is still P11.

6.6 Budget Pocket Quality Ranking

6.7 AP / EP / D / E / B Comparison

Chapter 7: Valuation Language Must Follow Business Control Points

Key Takeaway: The four companies occupy different tiers in terms of "new control points," and forcing the use of the same P/E language is equivalent to comparing four different types of errors. This chapter provides each company with a **valuation anchor + fair multiple range + adjustment rationale**.

7.1 Why the Same P/E Can No Longer Be Used

7.2 INTU's Valuation Language — Authority Owner Multiple

Valuation Anchor: Benchmark INTU against an authority owner of the caliber of FICO / Moody's / MSCI, rather than a software leader of the caliber of CRM / ADBE.

7.3 ADBE's Valuation Language — Franchise Defense Multiple

Valuation Anchor: ADBE should not be valued as "AI creative software"; it should be valued in layers based on "legacy franchise quality + AI nascent option."

Key Implication: Current 9.6x PE might be cheap, but the degree of cheapness depends on whether GenStudio can reach $2B by FY28 (management's target) — this is a CQ2 (whether the central axis of the split entity can lean towards the Option side).

7.4 ADSK's Valuation Language — Platform Migration Multiple

Valuation Anchor: ADSK should be valued by "strong workflow default franchise + P1 migration option," not purely by CAD seat.

7.5 PTC's Valuation Language — PEG + Control Layer Option

Valuation Anchor: PTC's Non-GAAP PE is already approximately equal to Owner PE, so 19.4x seems reasonable — but the market has not priced in the P1 control layer option.

7.6 Comparison Table of Four Companies' Valuation Languages

Chapter 8: Five Most Easily Misjudged Things

8.1 Misjudgment 1 — "INTU is just tax software; it will be disrupted by Direct File and AI tax filing"

Counter-Evidence: TurboTax accounts for 35-40% of INTU's total revenue; even if TT revenue decreases by 30%, the overall revenue impact is approximately -10-12%. CK +32% + SSE +18% can compensate within 2-3 years.

8.2 Misjudgment 2 — "ADBE's moat has been breached by AI"

Counter-evidence: DC +16%, GenStudio >$1B +30%, Firefly >$250M QoQ+75%. The narrative that "AI breaks Adobe" only holds true for 21% of revenue; 79% of revenue is benefiting or stable.

8.3 Misjudgment 3 — "ADSK's BIM mandate is a permanent moat"

Actual Judgment: The mandate is a high-probability moat (80%), not a certainty moat. In mature markets, Revit is stable at 60%+, while in emerging markets (China/India), Revit <40% has real room for alternatives.

8.4 Misjudgment 4 — "PTC is a patchwork of industrial software assets"

8.5 Misjudgment 5 — "All four companies are impacted by AI simultaneously"

Actual Implication: The net impact of AI on the four companies is in different directions. Whoever takes on more responsibility will benefit more from the revaluation of authority in the AI era.

Chapter 9: Kill Switch Monitoring Dashboard + Unified Tracking Metrics

9.1 Unified Kill Switch Table

9.2 Four Company Monitoring Calendar

9.3 Leading Indicators vs. Lagging Indicators

Chapter 10: Red Team — Where Our Framework Could Be Wrong

Red Team Objective: If the conclusions of this report are wrong, these are the five most likely paths for error. For each path, we provide the "what to observe for a reversal" trigger conditions.

10.1 Red Team 1 — Have We Overestimated INTU's Authority Monetization Capability?

Attack: INTU's I4 = 8/10 is based on IRS e-filer + Max Defend + Accuracy Guarantee. However, Max Defend is only an $80 add-on package, and it has not truly priced authority as its core business. Compared to MCO / FICO, its authority monetization depth is still an order of magnitude lower.

Rebuttal: This report acknowledges that INTU's authority does not yet account for a primary proportion of revenue like FICO/MCO. However, the significance of I4 lies in its **structural barrier** rather than **direct revenue**. INTU's IRS e-filer status grants it **regulatory authorization** in the question of "who can be the tax filing entry point," which is a structural advantage pure software companies cannot replace — it manifests in pricing as **customer stickiness** rather than direct revenue. This is more akin to Visa's interchange — not the primary revenue, but the cornerstone of the entire ecosystem.

Reversal Trigger: If the IRS lowers the e-filer authorization threshold to a point where any LLM provider can pass → INTU's authority transforms from structural to ordinary → I4 drops from 8 to 6 → Total score drops by 8-10 points → P4 downgrade.

10.2 Red Team 2 — Will ADBE Reclaim the Entry Point?

Attack: If the Firefly model marketplace strategy succeeds, ADBE could reposition ChatGPT/Midjourney as **pipelines** rather than **competitors** within the "professional creative workflow." This would imply that ADBE's I8 = 2.5/5 is an **underestimation**.

Rebuttal: This report acknowledges that the Firefly strategy is sound, but it's necessary to differentiate between "professional Pro workflow" and "SMB/Consumer initiated." Firefly can hold its ground within the Pro segment, but consumer-level users are already intercepted by ChatGPT/Canva **before opening Photoshop**. I8 = 2.5/5 is company-wide weighted, where the Pro side might be 4/5, and the Consumer side might be 1/5.

Reversal Trigger: If Canva MAU starts to decline from 265M + Express monthly active users exceed 200M + Midjourney users publicly report "it's more convenient to use via Adobe" → I8 might rebound to 3.5/5 → Total score +3~+5 → ADBE approaches ADSK levels.

10.3 Red Team 3 — Is ADSK's Workflow Default Actually Unstable?

Attack: This report assigns ADSK an I8 = 4/5, which is higher than ADBE's 2.5/5 and PTC's 4/5. However, AutoCAD drafting faces a real AI replacement threat, discussions around Revit's IFC open standard are intensifying, and Fusion lags PTC in MFG market share. ADSK's default appeal might stem from quantity rather than quality — multiple products layered together look good, but each line is fragile.

Rebuttal: This report acknowledges the risk of "default appeal looks good but each leg is fragile," but ADSK's I8 score primarily comes from:

1. Professional first-click — Engineers still start with Revit/AutoCAD/Fusion/ACC, not ChatGPT
2. Project-level institutional lock — BIM collaboration at the project level is a team decision, not an individual one, making it difficult for AI to replace
3. APS/Assistant native integration — The Assistant is embedded in Revit, not an external add-on; if the Assistant succeeds, I8 will rise further

Reversal Trigger: If Autodesk Assistant remains in Tech Preview after 18 months + APS call volume does not pick up + AutoCAD seat YoY consecutively drops -5% for 2 quarters → I8 falls from 4 to 2.5 → Total score -5 → ADSK drops back to ADBE's tier (67-68).

10.4 Red Team 4 — Is PTC still not P1 enough?

Attack: This report assigns PTC P1 Primary Archetype + P11 Secondary Archetype, but the report itself admits that 38% of the CAD family still has P11 attributes. 38% is not an insignificant number — if Creo cannot transition to Onshape, PTC might forever be stuck as P11 + Clean Owner Economics, rather than a true P1.

Rebuttal: This report acknowledges this is a true binary path. Currently, Onshape ARR is approximately $90-120M vs Creo's ~$600-700M, so Onshape's share is still small. However, Onshape's user profile (startup / education / new team) represents future incremental growth, while Creo represents existing users — even if Creo remains stable, Onshape's growth is sufficient to gradually increase the P1 attributes of the CAD family.

Reversal Trigger: Onshape ARR in 2027 remains <$150M + Creo YoY turns negative → Confirms no transition on the CAD side → P1 weighting downgraded → PTC total score -2~-4.

10.5 Red Team 5 — Correct Structure Does Not Equal Correct Odds

Attack: This report ranks business quality as INTU > PTC > ADSK > ADBE, but this does not imply the same investment return ranking.

Rebuttal: This report's business quality ranking is not investment advice, but rather a structural observation. Each company's actual return depends on the magnitude of (Current PE - Fair PE) / Current PE:

Conclusion: Business Quality #1 (INTU) and Business Quality #4 (ADBE) may both offer the best odds, as the former is undervalued for its authority, and the latter for its advantageous businesses within its fragmented structure. While PTC and ADSK have good structural quality, their PEs already reflect this. Business quality ranking + odds are not necessarily in the same order — this is the most important distinction between this report and single-dimension analysis.

10.6 Five Conclusions That Still Hold After the Red Team Challenge

1. INTU's structural advantage across the I4 + I7 + I8 three grids remains unshaken by the red team — even if the IRS e-filer threshold is lowered, the relative height of authority L3+L4 persists.
2. ADBE is a fragmented entity rather than a single-line decline — this diagnosis is evidenced by the six business line AIAS heatmap.
3. ADSK is more stable than ADBE — because I8 = 4 and I8 = 2.5 represent differences in business structure, not valuation techniques.
4. PTC's Clean Owner Economics remain unshaken by the red team — SBC 4.8% is hard data.
5. Business quality ranking ≠ odds ranking — this is a dimension that requires independent judgment, which this report has discussed separately in Chapter 7.

The first half (Ch 1-10) provides the overall structure and cross-sectional comparisons. The second half (Ch 11-16) conducts deep dives into mechanisms: for each company, a single most severely undervalued business line / structural mechanism is selected for a pinpoint penetration analysis.

Deep dive rationale: Market misjudgments are usually not "completely wrong" but rather occur when a specific detail is overlooked, leading to overall valuation deviation. This chapter specifically identifies that line for each company, providing quantifiable evidence and the magnitude of its impact.

Chapter 11: INTU Mechanism Deep Dive — Credit Karma is Already "$5.7B in Hidden Value" Yet the Market Still Discounts It Based on Mailchimp

11.1 Market Confusion: $12B Was Not CK's Acquisition Price

First, let's correct a widespread misconception — the "$12B mistake".

Key Observation: The market remembers the $12B lesson from Mailchimp and misapplies it to CK, severely discounting CK's true contribution.

11.2 Credit Karma's Three-Driver Lead Gen Model

CK's monetization model does not directly come from its 150M+ registered users, but rather from a B2B advertising/lead generation model.

Data Anchor Points:

• CK Contribution: FY25 $2.3B, +32%
• Q1 FY26 +27%, Q2 FY26 +23%
• Interest Rate Sensitivity: FY23-24 suppressed by high interest rates (fewer loans disbursed) → FY25 recovery indicates cyclicality ≠ structural failure

11.3 CK × TurboTax Data Crossover — The Flywheel Truly Begins

Core Insight: This is INTU's sole asymmetric advantage over QBO/Stripe/Plaid/Credit Sesame in I2 (cross-entity financial context) — tax filing is a complete declaration of annual income/assets, while other companies only have fragmented data.

Significance of 60% → 78% Accuracy Rate:

• 18pp improvement = 30% relative improvement
• Based on CK's FY25 $2.3B Lead Gen model, a 30% conversion rate improvement ≈ incremental $400-500M in revenue
• This incremental revenue can only be achieved through CK × TurboTax data crossover; CK alone cannot achieve it.

11.4 ProTax — The Most Unassuming Yet Most Important "Ecosystem Anchor"

ProTax FY25 revenue of $621M (only 3% of total), with +4% growth, is the smallest and slowest-growing of all INTU segments. However, it is the cornerstone of the QuickBooks ecosystem.

Key Quantification: ProTax's $621M revenue × 3% growth may seem mediocre, but its Lead Gen value is a significant source of low-cost customer acquisition for QBO's approximately $11B in revenue. If ProTax were to be discontinued, QBO's S&M costs would increase by at least 5-8%.

Why AI Cannot Bypass This Layer:

1. AI can replace simple tax filing, but cannot gain the professional trust of accountants
2. Accountants' tool choices are based on years of training + client migration costs, not "whose model is best"
3. INTU's launch of Intuit Accountant Suite (AI-native accountant platform) in Dec 2025 indicates management's proactive defense of this layer

11.5 INTU's Mechanism Synthesis — Why P4 Instead of P11

Comparison: ADBE/ADSK/PTC cover at most 1 to 1.5 of the three segments (ADBE's I8 creative workflow + partial I4 IP indemnification; ADSK's I8 workflow default; PTC's partial I4 TÜV). Only INTU possesses all three segments, which is the structural definition of P4.

Chapter 12: ADBE Mechanism Deep Dive — Document Cloud is the Most Severely Underestimated "Second Curve"

12.1 Why DC Deserves More Attention Than CC

When the market discusses ADBE, 80% of the time is spent on Photoshop / Illustrator / Firefly, but Document Cloud is ADBE's business line with the fastest growth + deepest moat.

Key Data:

• DC +16% YoY, Business Pro Subscriptions
• Nearly 50% of enterprise ETLA upgraded to AI version
• Acrobat AI Unique Advantages: Clickable Citations + Privacy Assurance (documents remain within the enterprise)

12.2 PDF ISO Standard — A Harder Moat Than Photoshop

Core Insight: PSD (Photoshop format) is an industry convention, while PDF is an ISO international standard. These are two magnitudes of moats.

Quantification:

• Approximately 20 billion documents are generated globally in PDF format daily
• Adobe Acrobat accounts for >60% of the enterprise PDF editor market
• No LLM or AI-native alternative possesses global ISO standard certification + local encryption + government compatibility

12.3 Differences Between Acrobat AI and General LLMs

Structural Implications: Enterprise legal/compliance/HR PDF workflows cannot migrate to ChatGPT, because these documents cannot leave the enterprise firewall. Acrobat AI's "clickable citations" are not a nice-to-have, but a necessity in legal scenarios — judges see "AI says contract Y is on page 15, line Z" rather than "AI said Y".

12.4 GenStudio — A Substantial Leap in Enterprise Content Governance

GenStudio is ADBE's most aggressive attempt to migrate to the I2+I3+I4 control layer.

Why GenStudio Is Hard to Replace:

1. Requires enterprise historical creative data — Midjourney/Runway generates "generic" creative, GenStudio generates creative "for this brand"
2. Requires workflow integration — Brand approval + legal compliance + localization are workflows, not features
3. Requires Top 50 client relationships — 90% adoption rate indicates it has been selected within enterprises

Implications for Valuation: If GenStudio scales from the current $1B ARR to $2B ARR by FY28, Adobe's label would shift from "creative tool company" to "AI-era enterprise content governance platform". A corresponding market P/E from 9.6x to 15-18x would be reasonable, implying a stock price from $256 to $400-480, with an upside of 56-87%.

12.5 Comparison of "Three Transformations" — Adobe is Undergoing Its Most Difficult One

Key Difference: In the first two transformations, Adobe's value remained at the same layer (tool layer), merely changing packaging. The third transformation requires Adobe's value to migrate to the governance layer — this is a leap "from selling tools to selling standards," a feat historically accomplished by only a handful of companies (Visa/Mastercard/FICO/Moody's).

Conditions Adobe Possesses:

• Cash flow support from legacy businesses CC Pro + DC: $13B+ annual revenue, $5B+ FCF
• Enterprise relationships: 90% of Top 50 have adopted AI-first
• Model aggregation capability: Firefly model supermarket strategy
• Legal IP guarantee: Firefly 100% legally trained commitment + IP compensation

Risks:

• Firefly training data controversy
• CC Consumer's continuous bleeding may exhaust patience before FY28
• DOJ $150M settlement, if escalated, could trigger pricing restructuring

Chapter 13: ADSK Mechanism Deep Dive — The $200M Noise Problem of Billing Transformation

13.1 Is the +18% reported revenue for FY26 fake?

The market's biggest confusion about ADSK is: FY26 officially reported revenue +18%, but management's FY27 guidance is +12-13%, with organic growth ~13%. Where does this gap come from?

13.2 Mechanical Nature of the Catch-up Effect — Not a Business Improvement in Figures

13.3 FY27 is the Key Validation Year

Key Observation: ADSK's valuation is directly anchored to a set of FY27 numbers. This differs from ADBE/INTU/PTC — the other three's valuations focus on long-term structure, while ADSK's relies on a three-pronged triangulation of Revenue/Billings/FCF for the next 12 months.

13.4 The Dual-Caliber Dilemma of Owner Economics

SBC Convergence Evidence:

• FY22 12.6% → FY26 10.9% → FY27 guidance <10%
• PTC SBC/Rev 4.8%, Bentley 4.8% — suggesting industrial software SBC has the potential to converge to 5-7%
• However, ADSK PSU terms are tied to Revenue + FCF targets — as long as management continues to pursue growth, SBC will not be significantly compressed

Chapter Judgment: The truth lies between Standard and Owner, with an **implied CAGR of approximately 13-14%**, slightly higher than guidance — indicating that the market is **slightly** pricing in AI capability improvements and APS options.

13.5 ADSK's Workflow Default Mechanism — Why It Remains Stable in the AI Era

Core Difference vs ADBE: Most professional creative work (photos/videos/documents) is an **individual decision**, so Canva can penetrate at a single point. Professional engineering (BIM/projects) is a **team collaboration decision**, so AI-native tools need to **replace the entire team's workflow** to shake Revit — which is a challenge of a different magnitude.

Chapter 14: Deep Dive into PTC Mechanism — BMW Case Study + The True Cross-Product Value of Codebeamer × Windchill

14.1 Why BMW's News on 2026-04-01 is a Significant Signal

On 2026-04-01, BMW Group announced the enterprise-wide adoption of Codebeamer as the standard for requirements/testing/traceability. This may seem like ordinary news, but it is the most critical verifiable anchor for the PTC P1 narrative.

14.2 Codebeamer × Windchill Unification — True P1 Evidence

Neither Windchill alone nor Codebeamer alone constitutes a P1 control layer, but their unification creates a true governed execution layer for regulated discrete manufacturing clients.

Quantified Value: Each new vehicle model approval + ISO 26262 report in the automotive industry currently requires 6-12 weeks of manual work, reduced to 2-4 days after unification. A BMW-level enterprise with 6-10 major models per year × 5-10 person-months saved per model = annualized savings of $5-20M. This is not PTC's revenue, but **the economic reason why customers cannot switch**.

14.3 ServiceMax — The Most Overlooked I4 + I7 Contribution

ServiceMax is a Service Lifecycle Management (SLM) product acquired by PTC in 2022 for $1.46B. The market often considers it "a small SaaS patch," but it is PTC's only business line with **true exception ownership** (I7).

I7 Value-Add Mechanism:

• SLA Commitment — Contract-level response time commitment (e.g., 24 hours); if not met, PTC compensates the client (client-side compliance)
• Entitlement engine — Which failures are covered by warranty and which are not; ServiceMax directly determines this
• Service board governance — Approval / escalation / exception handling of service workflows

These are capabilities that ADBE/ADSK completely **lack**.

14.4 Purest Owner Economics — Why PTC is the Only One Among Four

Direct Implications for Valuation: When investors see PTC's 19.4x P/E and ADSK's 19x P/E, **these two figures are not comparable**. ADSK's 19x is **artificially inflated**; its actual Owner P/E is 37x; PTC's 19x is already its Owner P/E. This is a frequently overlooked valuation misinterpretation.

Chapter 15: Comparison of Four Companies' Flywheel Models — Structural Enhancement vs. Stall in the AI Era

The paradigm shift framework not only examines **static control points** but also **dynamic flywheels**. Whether each action of a company strengthens the next I2/I3/I4 determines if it is undergoing **structural enhancement** or **structural stall**.

15.1 INTU's Flywheel — The Most Complete Data Loop

15.2 ADBE's Flywheel — Split Flywheel (CC Pro vs Consumer)

15.3 ADSK's Flywheel — Platform Uplift Prototype

Status: Flywheel is starting but not closed-loop — Assistant is still Tech Preview, Flex token usage data not disclosed.

15.4 PTC's Flywheel — Most Complete Governed Execution

15.5 Four Companies' Flywheel Strength Comparison

Chapter 16: Valuation Recalculation — Establishing a Reasonable Price Range for the Four Companies Based on This Framework

This chapter does not provide a precise target price, but rather uses the structural framework of this report to give a reasonable price range, allowing readers to judge the odds themselves.

16.1 INTU Valuation Recalculation

Base Assumptions:

• FY26-28 Revenue CAGR 15-16% (SSE +18% / Consumer +10% / CK +25% / ProTax +3%)
• FY28 Revenue approx. $29B
• FCF Margin maintained at 32-33% → FY28 FCF ~$9.5B
• WACC 10.5%, Terminal Growth Rate 3%

Reasonable P/E Derivation:

16.2 ADBE Valuation Recalculation

Base Assumptions:

• Segment-Weighted Growth Rate FY26-28 approx. 11-13%
• FY28 Revenue approx. $31B
• FCF Margin 33-35%, FY28 FCF ~$11B

Tiered Reasonable P/E:

FY28 Fair EV = $11B × 14.6 = $161B vs Current $105B = +53% Upside

Risk: If GenStudio stalls at $1B, EC/Firefly weight is reduced, weighted P/E compresses to 12x → Fair EV $132B vs Current $105B = +26% Upside.

16.3 ADSK Valuation Recalculation

Two Scenarios:

Probability-Weighted: $246 × 0.4 + $367 × 0.4 + $441 × 0.2 = $332 (+34% vs $247.57)

16.4 PTC Valuation Recalculation

Two Scenarios:

Probability-Weighted: $157 × 0.5 + $220 × 0.35 + $280 × 0.15 = $198 (+41% vs $140.60)

16.5 Summary of Four Companies' Valuation Recalculations

Key Observation: All four companies show positive upside (+34% ~ +52%). This is because the valuation anchor of this framework is structural control points, not the current market's financial valuation. The market is currently systemically underpricing these four companies, but the degree and path of underpricing vary:

• INTU: Authority premium is not priced in
• ADBE: Weighted advantage businesses within the fragmented entity are undervalued
• ADSK: P11→P1 migration option is not priced in
• PTC: Owner Economics cleanliness is not priced in

16.6 Why the Four Companies Cannot Be Directly Weighted and Bought — Correlation Considerations

Meaning: The four companies have a high overlap in systemic risks (AI impact / SaaS valuations / interest rates), so an "equal-weighted four-stock" portfolio cannot lower beta but can diversify idiosyncratic risk. A more reasonable portfolio is weighted by odds, rather than by business quality:

• INTU 35% (Optimal Portfolio)
• ADBE 30% (High odds but volatile)
• PTC 20% (Stable structure)
• ADSK 15% (No further additions before FY27 validation)

Chapter 17: Conclusion

17.1 How This Report Truly Differs from "AI's Left Hand and Right Hand"

When these two reports are put together, the core difference is not in the wording, but that the map has changed. "AI's Left Hand and Right Hand" is more like an action report under an old framework: it reminds investors which names seem to have a story, but that position sizing and timing should not be rushed. When writing this report, we shifted to a more fundamental question: instead of first asking who most resembles an AI beneficiary stock, we first asked which type of control point each of the four companies occupies. Once the question changes, the ranking, risk language, and valuation language will diverge accordingly.

Precisely because the question has changed, the rankings given by the two reports will not be entirely consistent. The core tension in the old report was AI's left-hand and right-hand rewrite of the software value chain, making it more conservative on PTC and ADSK; the core tension in this report is the reordering of control points, responsibility layers, and default entry points, thus bringing these two companies back to the center of structural discussion. The difference here is not simply bullish or bearish, but that the definition of the research subject has changed.

17.2 If the Conclusions of the Two Reports Conflict, Which One Should Take Precedence?

If the question is simply which report is more suitable as the main framework currently, we would prioritize this one. The reason is not that it is newer, but that it breaks down the four companies in more detail: INTU is no longer just a software leader but an authority owner; PTC is not just industrial software but an execution control layer with cleaner owner economics; ADSK is not a pure toolbox but a sample migrating from P11 to P1; and ADBE is certainly not a monolithic AI loser, but a collection of six business lines with diverging fates. Only by separating these four types of assets first does subsequent valuation comparison become meaningful.

However, this does not mean the old report is invalid. While "AI's Left Hand and Right Hand" has an older framework, it is more cautious at the action layer, and more sensitive to path risk, realization pace, market sentiment, and disclosure quality. If this report is understood as the steering wheel, the old report is more like the brakes. What is truly useful is not to choose between the two, but to clarify that they are each responsible for different levels.

17.3 A More Reasonable Comprehensive Judgment When Considering the Four Companies Together

17.4 How to Truly Use Both Reports Together for Investment Decisions

A more reasonable approach is not to dwell on "who is right or wrong," but to let each report do what it does best. This report is responsible for answering what these four companies truly are; the old report is responsible for reminding investors that even if the direction is right, the timing cannot be wrong. The former determines the map, the latter determines the speed.

Applying this combined approach at the action level, a more reasonable ranking is: INTU > ADBE ≈ PTC > ADSK. This ranking does not imply that ADBE and PTC are of the same quality, but rather that one has higher odds and the other has a cleaner structure; one is more like a value bet position, and the other is more like a research-upgraded position. As for ADSK, the more appropriate positioning is still "re-evaluated in terms of framework, but action still awaits validation."

Final point: If one only reads the old report, these four companies would be interpreted as "four degrees of vulnerability to AI impact"; if one only reads this report, they would be interpreted as "four different control point assets." After truly combining both reports, the more stable approach is not to overturn the old report, but to acknowledge: the new map dictates where to look, and the old map reminds us not to drive too fast.

100BAGGERS.CLUB

© 2026 Investment Research Agent. All rights reserved.