VIA Science (A) Custom Case Solution & Analysis

Evidence Brief: VIA Science (A)

1. Financial Metrics

• Capital Raised: The firm secured 2 million dollars in seed funding and 8 million dollars in Series A funding [Para 8].
• Market Opportunity: Global utility companies spend approximately 10 billion dollars annually on transformer maintenance and replacement [Exhibit 4].
• Revenue Model: Initial engagements were structured as pilot projects ranging from 100,000 to 250,000 dollars per project [Para 12].
• Sales Cycle: Average time from initial contact to contract signing for utility clients exceeds 12 to 18 months [Para 15].

2. Operational Facts

• Core Technology: Proprietary AI engine combined with a distributed ledger (blockchain) to enable Trusted Analytics, allowing data to remain on-site while models train remotely [Para 4].
• Headcount: 25 employees, primarily data scientists and software engineers based in Montreal and Somerville [Para 10].
• Key Product: Focus on predicting high-value asset failure, specifically power transformers for large-scale utility providers [Para 14].
• Geography: Operations focused on North America, Western Europe, and Japan (specifically through partnerships with TEPCO) [Para 18].

3. Stakeholder Positions

• Colin Gounden (CEO): Advocates for a product-led strategy over a service-led consultancy to achieve scalability [Para 3].
• Utility Executives: Express high concern regarding data privacy and security, often refusing to move sensitive operational data to the cloud [Para 21].
• Data Science Team: Prefers working on diverse, complex problems across different industries rather than repetitive asset modeling [Para 24].
• Investors: Pressuring for a repeatable sales motion and clear path to 100 million dollars in annual recurring revenue [Para 26].

4. Information Gaps

• Customer Acquisition Cost (CAC): The case lacks specific data on the cost to acquire a single utility client beyond qualitative descriptions of long cycles.
• Churn Rate: No data provided on renewal rates after the initial pilot phase for early utility adopters.
• Competitor Pricing: Internal costs for utilities to build similar AI capabilities in-house are not quantified.

Strategic Analysis

1. Core Strategic Question

• Should VIA Science remain a broad-based AI consultancy or pivot to a specialized product company focusing on utility asset management?
• Can the blockchain-enabled Trusted Analytics platform provide enough differentiation to overcome the inertia of long utility sales cycles?

2. Structural Analysis

Jobs-to-be-Done: Utility managers do not want AI; they want to prevent catastrophic transformer failures without violating strict data residency regulations. VIA solves the conflict between data privacy and predictive accuracy.

Value Chain: The current bottleneck is the data preparation phase, which consumes 80 percent of project time. Moving to a standardized product shifts the burden of data cleaning to the client via the Trusted Analytics interface.

3. Strategic Options

Option A: Specialized Utility Product (Recommended)

• Rationale: Focuses resources on the transformer failure problem where the ROI is highest and the pain point is most acute.
• Trade-offs: High concentration risk; if the utility sector slows down, the company has no secondary revenue stream.
• Resources: Requires hiring industry-specific sales experts rather than generalist data scientists.

Option B: Horizontal Trusted Analytics Platform

• Rationale: Market the blockchain-AI integration as a security tool for any industry with sensitive data (Healthcare, Defense).
• Trade-offs: Dilutes the brand and forces the company to compete with much larger software providers.
• Resources: Massive increase in marketing and general software engineering spend.

4. Preliminary Recommendation

VIA should commit to the utility vertical. The unit economics of a 10 billion dollar maintenance market provide a clear path to scale. By standardizing the transformer model, VIA can reduce the sales cycle and move from custom projects to a recurring software license model.

Implementation Roadmap

1. Critical Path

• Month 1-3: Standardize the Data Schema. Create a fixed template for utility transformer data to eliminate custom engineering for every new client.
• Month 4-6: Transition current pilot clients to the Trusted Analytics (TA) platform. This proves the technology works in a live, distributed environment.
• Month 7-12: Hire three senior sales directors with existing relationships at top-tier North American utilities.

2. Key Constraints

• Regulatory Inertia: Utility commissions may be slow to approve spending on AI-driven maintenance versus traditional physical inspections.
• Talent Retention: Data scientists may leave if the work becomes too repetitive within a single vertical.

3. Risk-Adjusted Implementation Strategy

To mitigate the long sales cycle, the firm must implement a land and expand strategy. Start with a low-cost, 90-day diagnostic on a small subset of transformers (10-20 units) to demonstrate accuracy before pitching a full-fleet deployment. This reduces the initial capital hurdle for the client and shortens the approval process.

Executive Review and BLUF

1. BLUF

VIA Science must pivot from an AI consultancy to a specialized product firm targeting the utility sector. The current service-heavy model is not scalable and will exhaust capital before reaching profitability. The Trusted Analytics platform solves the primary barrier to entry—data privacy—giving VIA a three-year window to dominate the 10 billion dollar transformer maintenance market. Success requires standardizing the data model and replacing generalist data scientists with industry-specific sales leadership. Approved for leadership review.

2. Dangerous Assumption

The most dangerous premise is that utility companies will value data privacy (via blockchain) enough to pay a premium or move faster. If the delay in adoption is caused by budget cycles rather than security concerns, the Trusted Analytics technology is a solution to a secondary problem.

3. Unaddressed Risks

• Execution Risk (High): The transition from custom code to a standardized product often fails when the first three clients have slightly different data requirements, leading back to service-led creep.
• Market Risk (Moderate): A major cloud provider (AWS or Google) could release a secure, federated learning tool that mimics the privacy benefits of VIA’s blockchain approach, erasing their technical advantage.

4. Unconsidered Alternative

The team failed to consider a joint venture with a major hardware manufacturer (like GE or Siemens). Instead of selling software to utilities, VIA could embed its predictive models directly into the sensors of new transformers, capturing the value at the point of manufacture and bypassing the long utility sales cycle entirely.

5. MECE Analysis of Market Entry

• Direct Sales: Focus on top 50 global utilities.
• Channel Partners: Integrate with existing grid management software providers.
• OEM Integration: Embed AI in the hardware manufacturing process.