The Software Rebel: How Innovaccer Defied Silicon Valley's Anti-Services Doctrine and Won

Unconventional contrarian thinking is the path to a $100B company

With the AI revolution, one of the things top of mind for people is how to marry software and services. We often forget though - that this is how software itself was born – think massive ACV solutions sales from IBM, think large defense contractors, think private satellite network at Walmart. As we approach the software + services revolution, it’s important to go back and look at some of the lessons from history.

In the realm of Silicon Valley, a hush-hush secret prevails—services lack scalability, software reigns supreme, and minimizing human involvement is the key to success. It was against this backdrop that I made the initial investment in InnovAccer. Today, witnessing InnovAccer's remarkable growth into a $3.2B enterprise, poised to reach a $100B valuation, I feel compelled to clarify what truly drives success in enterprise software.

In this age of AI and all the hype around innovation, InnovAccer's approach is like a breath of fresh air. To get a real understanding of how their model works and why it could be the next big thing in enterprise software, we need to look at what the most successful companies in the space have done. Forget about the simplified stories and focus on the real strategies that helped them build their empires.

The Epic Blueprint: Lessons from Healthcare's Quiet Giant

In 1979, when Judy Faulkner started Epic Systems, she broke every rule that venture capitalists now hold sacred:

• Headquartered in Wisconsin, not Silicon Valley

• Bootstrap funded, no VC money

• Heavy services component

• Focused on unsexy healthcare back-office operations

• Didn't try to disrupt or move fast and break things

The metrics would make VCs cringe:

• Implementation revenue: 35% of total

• Average implementation time: 18 months

• Average initial contract: $10M

• Services margins: 30-40%

But look at the results:

• 250 million patient records

• $3.8B in annual revenue

• 99% customer retention

• 31% market share in large hospitals

When I first met Abhinav, Kanav and Sandy in 2014, they had studied Epic's model deeply. They saw what most missed: Epic's "services problem" was actually their moat. Every implementation taught them something about how hospitals really worked—knowledge that couldn't be gained any other way.

Veeva: Taking the Services-First Model to Life Sciences

Peter Gassner's creation of Veeva provided another crucial piece of the puzzle – they are the lifeblood of massive pharma companies. Raised only $4M over their entire lifetime, and valued at over $30B now. The first version of their software was build atop Salesforce. Were they a software or a services company? Their early metrics mirror what we'd later see at Innovaccer:

Veeva's Early Days:

• 70% services revenue

• 15-20% services margin

• $380K customer acquisition cost

• 4-6 month implementation timeline

But watch the progression:

• Year 1-2: 70% services / 30% software

• Year 3-4: 45% services / 55% software

• Year 5-6: 30% services / 70% software

• Today: 20% services / 80% software

The customer economics tell the real story:

• Initial contract value: $250K

• Average value after 3 years: $1.2M

• Customer retention: 98%

• Gross margin progression: 45% → 73%

Marry all of this with a CEO who was focused on efficient growth from the beginning, and laser focused on delivering customer value - and you get one of the highest alpha investments of all time. The company has created $30B in market cap from only $4M in investment from Emergence Capital.

Salesforce: The Hidden Services Story Silicon Valley Forgot

When people talk about Salesforce today, they paint a picture of pure SaaS perfection. But having heard their story from the trenches, I can tell you the real story was very different.

The Hidden Metrics (2000-2004):

• Professional services revenue: 30-35%

• Implementation partners: 150+

• Average enterprise deployment: 4-6 months

• Customer success team ratio: 1:8 customers

What people forget is that moving to the cloud was controversial then. Every enterprise deployment required:

• Change management consulting

• Custom integrations

• Workflow redesign

• Training programs

• Security consulting

The real genius of Salesforce wasn't eliminating services—it was productizing them:

• 2000: 40% of implementations required custom code

• 2005: 25% required custom code

• 2010: 15% required custom code

• Today: 5% require custom code

But they never eliminated services completely. Instead, they:

• Built a massive partner ecosystem

• Created certification programs

• Standardized implementation methodologies

• Turned services into a strategic advantage

Current Metrics:

• Partner ecosystem revenue: $15B+

• Implementation success rate: 92%

• Customer retention: 92%

• Net revenue retention: 120%

Palantir: The FDE Revolution

Palantir has been misunderstood for most of their life. However, they are also a poster child of the AI revolution. Its stock has grown 170% in 2024 alone, and they are now one of the most highly valued AI companies. Again - massive ACVs, and large sticky and niche use cases.

Their Forward Deployed Engineer (FDE) approach was getting criticized for exactly the same reasons people would later criticize our approach at Innovaccer.

Palantir's Controversial Metrics: 2010:

• 80% services revenue

• FDE cost: $400K+ per engineer

• Implementation time: 180+ days

• Gross margins: 40%

The conventional wisdom said this would never scale. But look what actually happened:

2023:

• 25% services revenue

• Implementation time: 45 days

• Gross margins: 70%+

• Market cap: $25B+

The key insight: FDEs weren't just implementing software—they were:

• Gathering intelligence about customer needs

• Identifying automation opportunities

• Building deep domain expertise

• Creating repeatable solutions

Joe Lonsdale recently wrote about this misunderstood insight. This was exactly what we'd see later at Innovaccer, but in healthcare.

My own experience at Capillary

During my time at Capillary, I witnessed a groundbreaking rodeo. We transformed how retailers interacted with their customers, but we soon realized that the true value lay in the insights derived from the treasure trove of data.

However, each retailer presented unique challenges, making it difficult to comprehend their business nuances. Our brilliant solution was to hire teams of Business Analysts. These talented individuals helped retailers interpret their data, enhancing the value of our solution tenfold.

Customers no longer simply purchased software; they gained insights that transformed their businesses. These insights empowered our customer champions, boosting their credibility within their management and solidifying our role in their success.

Additionally, our company became a learning academy. Engineers and Product Managers could collaborate with analysts, gaining insights into building an adaptable and customizable product that met the needs of diverse enterprise customers.

We pursued this strategy aggressively. With each iteration of the product, we freed up our analysts to focus on higher-value tasks, enhancing our flywheel.

While traditional wisdom advised against this approach, citing lower multiples for services companies and scrutinizing unit economics, we found that customers were experiencing real value. We were learning at an accelerated pace, and within a short time, the narrative began to change.

Innovaccer: Bringing It All Together

When Abhinav pitched me Innovaccer in 2014, I saw something others missed. They weren't just building another healthcare software company—they were taking the best elements of Epic, Veeva, Salesforce, and Palantir and adapting them for healthcare's unique challenges – which all added up in my head given what I had seen at Capillary:

From Epic:

• Deep healthcare domain expertise

• High-touch implementation model

• Focus on customer outcomes

• Long-term relationship building

From Veeva:

• Industry-specific platform approach

• Systematic services-to-software evolution

• Clear vertical focus

• Regulatory compliance expertise

From Salesforce:

• Cloud-first architecture

• Partner ecosystem development

• Standardized methodologies

• Training and certification programs

From Palantir:

• Forward deployed teams

• Knowledge capture systems

• Pattern recognition

• Systematic automation

But Innovaccer took our own model a lot further. They created the Forward Business Analyst (FBA) model, which combined technical expertise with deep healthcare business knowledge, with a relentless focus on productizing their knowledge.

Some metrics:

• Services Revenue went from 65% at start to 25%: surgically delivered to enhance the value of the largest contracts

• Gross Margins went from 45% to over 75%

• Implementation Cycles compressed from 6 months to under a month

• ACV went from $250K to over $3M

The key innovation? Innovaccer turned FBAs into a learning engine:

1. The Knowledge Flywheel

• Every FBA implementation captured in knowledge base

• ML models trained on implementation patterns

• Automated solution recommendations

• Continuous feedback loops

2. Value Creation For customers with full FBA engagement:

• 3x faster ROI realization

• 4x higher value creation

• 92% reduction in data mapping errors

• 85% reduction in implementation time

Running the playbook for AI

InnovAccer is reapplying their proven strategies for their latest AI products:

• Accelerate time to value from 9 months to under 3 months with the help of feedback loops.

• Double customer return on investment (ROI) within the first quarter.

• Achieve 95% or higher accuracy with AI solutions powered by human input.

FBAs revolutionize AI workflows by:

• Training AI models using real healthcare data.

• Customizing AI solutions for specific use cases.

• Prioritizing responsibility, compliance, and continuous improvement.

These enhancements have already resulted in a 40% increase in model accuracy and an 85% reduction in compliance issues, leading to customers adopting their product three times faster.

However, InnovAccer is not unique in this approach. Leading AI giants also employ teams dedicated to fine-tuning, customizing, and developing solutions. While the flashy marketing campaigns capture our attention, the true work takes place behind the scenes, where these companies strive to improve margins in the long run.

The Real Enterprise Software Playbook for the AI Era

Based on Innovaccer's success, here's what actually works:

1. Embrace Services as a Strategic Weapon

First 12 Months:

• Target 60-70% services revenue

• Build domain expertise systematically

• Document everything

• Be truly customer obsessed

Metrics to Track:

• Pattern recognition rate

• Knowledge base growth

• Service efficiency gains

• Customer success correlation

2. Build the Learning Engine

Knowledge Capture:

• Map 100+ customer workflows

• Create standardized assessment tools

• Build internal knowledge base

• Establish feedback loops

Efficiency Targets:

• 20% reduction in implementation time quarterly

• 30% increase in reusable components

• 40% improvement in first-time resolution

• 50% reduction in custom code

3. Systematic Automation

Phase 1 (Months 1-18):

• Automate 20% of common tasks

• Build internal tools for service team

• Create template library

• Extensible architecture

• Standardize implementations

Phase 2 (Months 18-36):

• Automate 50% of common tasks

• Launch self-service components

• Deploy AI assistance

• Scale knowledge base

The Future of Enterprise Software

Based on what we've learned, here's what's coming:

1. The Return of Domain Expertise

• AI will commoditize generic software

• Industry-specific knowledge becomes more valuable

• Hybrid teams (human + AI) become the norm

• Services margins improve through AI augmentation

2. The New Enterprise Stack

• Core AI capabilities via API

• Industry-specific solutions built on top

• Human expertise augmented by AI

• Continuous learning loops

3. The Services Renaissance

• High-value services become premium offerings

• AI augments but doesn't replace expertise

• Services margins improve through automation

• New hybrid business models emerge

Caveat Emptor

Don’t do this blindly. Consider these factors:

1. Customer Size: Is your target market composed of large enterprises with substantial ACVs? If you're aiming for smaller customers with annual contracts worth $10K or less, this strategy may not be suitable.

2. Learning Culture: Can you foster a culture that embraces continuous learning and rapid adaptation? You'll face significant pressure, so you and your team must be equipped to out-innovate a rapidly evolving customer base.

3. Strategic Focus: Carefully select the challenges you want to tackle and choose the appropriate strategies to address them.

A Call to Action for Founders

If you're building truly enterprise software today, here's what you need to know:

1. Reject the Pure Software Dogma

• Embrace services as strategic advantage

• Build learning loops into your company

• Focus on customer outcomes, not margins

• Think decades, not quarters

2. Build the Right Team

• Hire domain experts early

• Create knowledge capture systems

• Build hybrid capabilities

• Invest in services leadership

3. Focus on Learning Rate

• Measure knowledge acquisition

• Build feedback loops

• Document everything

• Convert insights to features

The Final Word

In the era of AI-driven enterprise software, the conventional wisdom of building software companies solely on technology is being challenged. The most successful companies in this domain will be those that strategically leverage services to gain a deep understanding of problems and develop solutions that truly address them. Innovaccer is a prime example of this, as their integration of services and software has positioned them to potentially become a $100B healthcare technology company.

This shift in perspective is crucial as AI transforms enterprise software. The future belongs to knowledge companies that employ software and AI to extend their expertise. Innovaccer has demonstrated this approach and serves as a model for the next generation of enterprise software companies.

The "services heresy," once considered controversial, is now recognized as the future of enterprise software. The question remains: are you ready to embrace this new reality?

This is the untold story of building real enterprise software companies, a far cry from the idealized narratives often presented at startup conferences. My experiences at Capillary and my early investment in Innovaccer have taught me valuable lessons about what truly works in this sector. Every founder needs to understand this paradigm shift to succeed in enterprise software.