Despite Massive GenAI Spend, ROI Remains Elusive — But Anna E. Molosky Says the Fix Is All About Sequencing

Enterprises are pouring billions into GenAI, yet many still struggle to prove real financial impact.

According to Anna E. Molosky, the challenge isn’t the technology — it’s the order of operations. Too many organizations leap straight into advanced, flashy AI use cases without establishing the operational backbone needed to produce dependable returns.

Molosky offers three pragmatic, ROI-first steps to shift AI from experimentation to measurable enterprise value.

1️⃣ Put Operational Automation First — Not Last

Executives often assume customer-facing GenAI initiatives (marketing content, sales intelligence, digital experiences) will generate the fastest wins.
But the data — and years of enterprise automation history — tell a different story.

Molosky emphasizes that the most reliable returns come from automating:

• Finance workflows

• HR transactions

• Shared services functions

• High-volume, rules-driven processes that don’t require GenAI at all

These areas have clear baselines, structured processes, and immediate cost takeout — making them ideal for rapid, verifiable ROI.

The evidence is compelling: AT&T’s enterprise-wide automation program cut 16.9 million manual minutes annually, delivered 20x ROI, and unlocked hundreds of millions in recurring value.

Molosky’s view:
If you want guaranteed ROI, automate the processes you already understand before trying to automate the ones you’re still redesigning.

2️⃣ Convert Shadow AI From a Liability Into a Strategic Advantage

Employees are already demonstrating GenAI’s value — often without organizational support.

• 90% use personal AI tools to accelerate work

• Only 40% have employer-provided, governed access

Molosky argues that this isn’t merely a compliance risk — it’s a roadmap.
Shadow AI activity reveals exactly which tasks employees find worth automating.

To turn this into advantage, leaders should:

1. Audit how employees already use AI

2. Identify high-frequency, high-impact patterns

3. Convert those into secure, enterprise-grade workflows

This approach leverages the best source of automation insights: the employees doing the work.

As Molosky puts it:
Your workforce has already drafted your first AI roadmap — you just haven’t captured it.

3️⃣ Align AI ROI Expectations With Enterprise Reality

One of the most common strategic errors in AI adoption is timeline compression.
Molosky underscores a simple but often ignored truth: enterprise-scale technology cannot mature in a six-month window.

Global deployments typically require 1–3 years, especially when they involve:

• Systems integration

• Data cleanup and harmonization

• Process redesign

• Organizational and global change management

• Governance, training, and risk controls

Short-term evaluation windows make healthy, in-progress initiatives appear “failed.”
Instead of treating the “95% failure rate” as a red flag, Molosky frames it as a maturity curve:
the top 5% are simply further along the adoption lifecycle, not fundamentally better.

Molosky’s ROI-First Formula for GenAI

To move GenAI from hype to results, enterprises should:

🔹 Shift early investment to back-office, high-certainty automation
🔹 Use employee-driven AI behavior to identify proven value opportunities
🔹 Set timelines that reflect actual enterprise transformation complexity

Follow the right sequence and GenAI stops being a gamble — it becomes an engine for predictable operational and financial performance.