thesis
Three Pillars: Tech/SaaS Application Lens
Adaptation of the Three Pillars framework for technology and SaaS companies where the baseline assumptions differ. Addresses how service-ization, data assetization, and predictive resilience apply when companies are already digital-native, data-rich, and subscription-based.
The Problem with Our Current Framing
The universal pillars assume a traditional company transformation journey:
| Pillar | Traditional Assumption | SaaS Reality |
|---|---|---|
| P1: Service-ization | "Move from transactions to relationships" | Already subscription; relationships are baseline |
| P2: Data Assetization | "Start treating data as an asset" | Swimming in data; challenge is extraction, not collection |
| P3: Predictive Resilience | "Build prediction capability" | Often have churn models; challenge is scale and action |
The risk: We sound naive talking to a SaaS CEO about "creating recurring revenue" or "starting to use your data."
The opportunity: SaaS companies have different versions of these challenges that are often underaddressed.
How the Pillars Actually Apply to Tech/SaaS
Pillar 1 for SaaS: From "Customer Success" to "Embedded Partnership"
Traditional P1: Create service relationships where none exist SaaS P1: Deepen existing relationships to create defensibility and expansion
The SaaS Service-ization Challenge
SaaS companies already have recurring relationshipsβthat's the model. But most are stuck at shallow engagement:
SAAS SERVICE MATURITY LEVELS
Level 1: SUPPORT (Reactive)
βββ Help desk responds to tickets
βββ Knowledge base exists
βββ Success = problems solved
βββ Reality: Most SaaS companies stop here
Level 2: SUCCESS (Proactive-ish)
βββ CSMs assigned to accounts
βββ Onboarding programs
βββ QBRs and check-ins
βββ Success = retention
βββ Reality: Resource-constrained; CSM ratios 1:50-200
Level 3: EMBEDDED PARTNERSHIP (AI-enabled)
βββ AI handles routine success at scale
βββ Humans focus on strategic accounts
βββ Customer operations improved by vendor
βββ Success = customer's business outcomes
βββ Reality: Where the moat is built
The Specific SaaS Challenges
Challenge 1: White-Glove at Startup Scale
- Enterprise customers expect high-touch
- Startup resources can't afford 1:20 CSM ratios
- AI enables enterprise experience without enterprise cost structure
Challenge 2: Product Commoditization
- Features get copied; UI converges
- Service/success becomes the differentiator
- "We help you succeed" beats "we have features"
Challenge 3: Expansion as the Growth Engine
- New logo acquisition gets harder/more expensive
- Net revenue retention >120% is the SaaS holy grail
- Expansion requires deep customer understanding
Challenge 4: Moving Upmarket
- SMB β Mid-Market β Enterprise requires service capability
- Can't hire your way to enterprise service
- AI enables punching above weight class
SaaS P1 Applications
| Application | Description | AI Enable |
|---|---|---|
| Scaled Customer Success | 1:many proactive engagement | AI handles routine; surfaces exceptions |
| Usage-Based Coaching | "You're not using X feature that would help with Y" | Pattern recognition across customer base |
| Proactive Risk Intervention | Engage before churn signals escalate | Real-time health scoring with auto-triggers |
| Outcome Tracking | Prove ROI in customer's terms | Automated value realization reporting |
| Strategic Account Intelligence | Deep prep for CSM conversations | AI synthesis of account history/context |
SaaS P1 Case Examples
Gainsight's Own Transformation:
- Built CS platform on own customer base
- AI-assisted health scoring and playbook triggers
- Result: Proved the model before selling it
HubSpot's Inbound Success:
- Content + tools + success coaching integrated
- Free tools create relationship; success deepens it
- Result: Industry-leading NRR for their segment
Figma's Community-Led Success:
- Product enables community; community enables success
- AI could enhance: personalized learning paths, usage coaching
- Result: Viral expansion within accounts
Pillar 2 for SaaS: From "Start Using Data" to "Create Platform Value"
Traditional P2: Recognize data as an asset; begin monetization SaaS P2: Extract maximum value from data you're already generating
The SaaS Data Challenge
SaaS companies generate massive data but often under-leverage it:
SAAS DATA VALUE PYRAMID
βββββββββββββββββββ
β PLATFORM/ β β Few achieve
β ECOSYSTEM β
β (Network effects)
ββββββββββ¬βββββββββ
ββββββββββ΄βββββββββ
β DATA PRODUCTS β β Emerging
β (Benchmarks, β
β insights) β
ββββββββββ¬βββββββββ
ββββββββββ΄βββββββββ
β PRODUCT β β Most stop here
β IMPROVEMENT β
β (Better UX) β
ββββββββββ¬βββββββββ
ββββββββββ΄βββββββββ
β BASIC β β Baseline
β ANALYTICS β
β (Dashboards) β
βββββββββββββββββββ
The Specific SaaS Challenges
Challenge 1: Feature Parity = No Moat
- Every competitor can copy features
- Data creates the moatβbut only if used
- Network effects from data aggregation are the ultimate defensibility
Challenge 2: Usage Intelligence Gap
- Tons of usage data; limited usage understanding
- "Which features drive conversion?" still hard to answer
- Product decisions based on intuition, not data
Challenge 3: Benchmark Opportunity
- "How do I compare to peers?" is universal customer question
- SaaS vendors have cross-customer visibility
- Massively underleveraged opportunity
Challenge 4: Platform Evolution
- Point solution β Platform β Ecosystem
- Data is the bridge to platform
- Without data strategy, stuck as point solution
SaaS P2 Applications
| Application | Description | Value Creation |
|---|---|---|
| Benchmark Products | "You vs. industry" insights | New revenue stream; customer stickiness |
| Usage-Based Pricing Intelligence | Optimize pricing from usage patterns | Revenue optimization; reduced churn |
| Product-Led Growth Optimization | Which free actions predict conversion? | CAC reduction; conversion improvement |
| Feature Investment Prioritization | Which features drive retention/expansion? | R&D efficiency; outcome-driven roadmap |
| Cross-Customer Intelligence | Patterns that predict success | Product improvement; success playbooks |
| Platform/API Data | Enable ecosystem around your data | Network effects; ecosystem lock-in |
SaaS P2 Case Examples
Salesforce's Data Cloud:
- Unified customer data across ecosystem
- Became the "customer data platform" play
- Result: Platform lock-in; ecosystem stickiness
Stripe's Sigma:
- Analytics product built on payment data
- Merchants analyze their own data; Stripe sees all
- Result: Intelligence advantage + revenue stream
Notion's Template Ecosystem:
- User-created content creates data moat
- Templates as data products
- Result: Network effects from content
LinkedIn's Talent Insights:
- Workforce data monetized as intelligence product
- Customers get benchmarks; LinkedIn gets data moat
- Result: $1B+ Talent Solutions business
Pillar 3 for SaaS: From "Build Prediction" to "Scale Intervention"
Traditional P3: Develop prediction capability where none exists SaaS P3: Act on predictions at scale; embed prediction in product
The SaaS Prediction Challenge
Most mature SaaS companies have churn prediction models. The problem isn't predictionβit's what happens after:
PREDICTION-TO-ACTION GAP
Prediction Action Outcome
βββββββββββ βββββββββββ βββββββββββ
β Churn βββββββββΊβ CSM ββββββββββΊβ Save β
β Model β β Review β β (maybe) β
β (good) β β (slow) β β β
βββββββββββ βββββββββββ βββββββββββ
β β β
β BOTTLENECK β
β β
β The prediction is fine. β
β The action doesn't scale. β
ββββββββββββββββββββββββββββββββββββββββ
The Specific SaaS Challenges
Challenge 1: Intervention Doesn't Scale
- Model predicts 500 at-risk accounts
- CSM team has capacity for 50 saves
- Result: Good prediction, poor prevention
Challenge 2: Product-Embedded vs. Bolt-On
- Analytics dashboard shows risks (bolt-on)
- vs. Product itself prevents risks (embedded)
- Embedded prevention is 10x more effective
Challenge 3: Beyond ChurnβFull Risk Portfolio
- Churn prediction is table stakes
- What about: feature adoption failure, support escalation, expansion timing, champion departure?
- Prediction capability underleveraged
Challenge 4: Infrastructure Resilience
- Business risks matter; infrastructure risks existential
- Downtime, security incidents, performance degradation
- Prediction for infrastructure health
SaaS P3 Applications
| Application | Description | Prevention Mechanism |
|---|---|---|
| Automated Health Interventions | Predict risk β Auto-trigger engagement | In-app messaging, email sequences, feature prompts |
| Product-Embedded Prevention | UX changes based on risk signals | Onboarding paths adapt; friction removed proactively |
| Champion Risk Detection | Key user disengagement or departure | Relationship diversification prompts |
| Expansion Timing Prediction | When is account ready for upsell? | Sales trigger vs. spam; right message, right time |
| Support Escalation Prevention | Predict ticket β solve before ticket | Proactive help; known issue notification |
| Infrastructure Prediction | Capacity, performance, security signals | Auto-scaling; preemptive maintenance |
SaaS P3 Case Examples
Pendo's Product-Led Intervention:
- In-app guides trigger based on behavior signals
- Prediction + immediate action in product
- Result: Prevention embedded, not bolted on
Amplitude's Behavioral Cohorts:
- Predict which user behaviors lead to retention
- Product teams optimize for predictive behaviors
- Result: Prevention through product design
PagerDuty's Operational Prediction:
- Predict incidents before they escalate
- Auto-remediation where possible
- Result: Prevention as the product itself
Alternative Framework Question
Are the Pillars the Right Lens for SaaS?
The three pillars adapt to SaaS, but there's an argument for a different primary framework:
Alternative: SaaS Growth Efficiency Framework
| Dimension | Question | AI Application |
|---|---|---|
| Acquisition Efficiency | How do we reduce CAC? | Lead scoring, PLG optimization, content/SEO |
| Activation Efficiency | How do we accelerate time-to-value? | Personalized onboarding, usage coaching |
| Retention Efficiency | How do we prevent churn at scale? | Health prediction, automated intervention |
| Expansion Efficiency | How do we grow existing accounts? | Expansion timing, cross-sell intelligence |
| Advocacy Efficiency | How do we scale word-of-mouth? | Champion identification, referral optimization |
This is essentially AARRR (Pirate Metrics) + AI.
Alternative: SaaS Moat Framework
| Moat Type | Description | AI Enable |
|---|---|---|
| Network Effects | More users = more value | Data aggregation, cross-customer intelligence |
| Switching Costs | Pain to leave | Deep integration, workflow embedding |
| Data Advantages | Unique insights | Benchmark products, intelligence layer |
| Brand/Trust | Reputation | Success stories at scale, outcome proof |
Recommendation: Pillars + SaaS Adaptation
The three pillars remain useful as strategic categories, but for SaaS:
- Reframe the baseline β Acknowledge they're not starting from zero
- Emphasize scale and defensibility β The SaaS challenges are about scaling and moat-building
- Add SaaS-specific applications β Different use cases than traditional industries
- Consider alternative frameworks β Growth efficiency may resonate more with some SaaS leaders
SaaS Pillar Fit Diagnostic Adaptation
When evaluating a SaaS company, different signals matter:
| Factor | Traditional Signal | SaaS Signal |
|---|---|---|
| P1 Priority | No recurring revenue | NRR < 100%; moving upmarket; CSM overwhelmed |
| P2 Priority | Data not used | Feature parity with competitors; no benchmark product; platform aspiration |
| P3 Priority | Firefighting mode | Churn prediction exists but intervention doesn't scale; infrastructure incidents |
SaaS-Specific Discovery Questions
For P1 (Service/Success at Scale):
- "What's your CSM-to-account ratio? How do you prioritize?"
- "What's your NRR? Where does expansion come from?"
- "How do you handle enterprise expectations with startup resources?"
- "When a customer churns, when did you actually lose them?"
For P2 (Data/Platform Value):
- "How do you know which features drive retention vs. just get used?"
- "Have customers ever asked 'how do I compare to others'?"
- "What's your platform strategy? Where does data fit?"
- "What do you know about your customers that competitors can't know?"
For P3 (Prediction/Prevention at Scale):
- "You have a churn modelβwhat happens after it flags an account?"
- "When did your last save conversation happen relative to the prediction?"
- "What risks beyond churn are you tracking?"
- "How does your product itself prevent failure vs. your team?"
SaaS Case Studies to Develop
| Category | Company | Pillar | Application |
|---|---|---|---|
| Customer Success | Gainsight | P1 | AI-assisted scaled success |
| Customer Success | HubSpot | P1 | Content + tools + success integration |
| Data/Platform | Salesforce | P2 | Data Cloud platform play |
| Data/Platform | Stripe | P2 | Sigma analytics product |
| Data/Platform | P2 | Talent Insights monetization | |
| Prediction/Action | Pendo | P3 | Product-embedded intervention |
| Prediction/Action | PagerDuty | P3 | Prevention as the product |
| Prediction/Action | Amplitude | P3 | Behavioral prediction for product design |
Implications for MPDrexel Positioning
Who Is Our SaaS Client?
Not every SaaS company needs us. The sweet spot:
Good fit:
- Series B-D stage (growth mode, not yet enterprise resources)
- $10-100M ARR (enough scale for AI to matter, not enough for internal teams)
- Vertical SaaS (domain expertise matters; not pure horizontal)
- Moving upmarket (SMB β Mid-Market β Enterprise)
- Platform aspirations (not content with point solution)
Less good fit:
- Pre-PMF (not our problem)
- Post-IPO enterprise (have internal teams)
- Horizontal infrastructure (different game entirely)
- Developer tools (community-driven; different model)
How We Position to SaaS
Don't say: "We help you start using your data / create service relationships / build prediction"
Do say:
- "We help you scale success without scaling headcount"
- "We help you build data moat before competitors catch up"
- "We help you close the prediction-to-action gap"
- "We help you do enterprise service at startup cost"
Version History
| Version | Date | Changes |
|---|---|---|
| 1.0 | 2026-01-03 | Initial SaaS lens analysis |