Home / Blog

Where to Invest in AI Tools When You’re Ready to Expand Use

Learn how to decide where to invest in AI tools next using cross-team adoption data, maturity benchmarks and productivity signals instead of gut-feel scoring.

Sarah Altemus

By Sarah Altemus

A stacked pyramid of colorful blocks labeled "LOW", "MODERATE", and "HIGH", topped with star shapes beneath the text "ROI" to represent AI tool investment levels.
Table of contents

Your team wants more AI, and someone in the room is pushing hardest for the next tool, license or pilot. But volume isn’t a strategy. If you lead operations, finance or the C-suite, you already have better inputs than opinion: cross-team adoption data, maturity benchmarks and productivity signals sitting inside your work intelligence platform. Here’s how to use that data to decide where your next AI dollar should go.

Why the loudest voice isn’t your best AI investment guide

Most AI prioritization advice starts with a scoring rubric built on opinion; this can be a workshop, a survey or even a gut-check from whoever asks first. Deloitte’s 2026 enterprise survey found only about a third of organizations use AI to deeply transform key processes, while a similar share remain stuck at a surface level with little real change.

Behavioral data closes the gap between meaningful use and surface-level exploration. ActivTrak’s assessment of more than 443 million hours of work activity shows which teams actually use AI, how deeply they use it and the impact that this activity had on work output.

How to read cross-team AI adoption data

To read and understand cross-team AI adoption data, leaders need to start with usage depth. Look at how many hours each team spends in AI tools relative to total work hours, how consistently they return month over month and how many platforms they’re juggling.

Most organizations aren’t consolidating around a few trusted tools; They’re expanding across an average of seven AI platforms per company. Making comparisons is more difficult when there is no behavioral data to anchor them.

Spot high-usage, high-friction teams first

The teams worth funding first aren’t always the ones using AI the most. Look for teams combining high usage with high friction; This will show up as repeated requests, manual workarounds or unsanctioned tools outside your approved stack. That combination signals unmet demand your current investment isn’t meeting.

Closing this visibility gap matters, because most organizations can see that AI is used; Few can see whether it changes how work gets done.

Benchmark your AI maturity by department

Adoption data tells you who’s using AI. Maturity benchmarks tell you how well they use AI. ActivTrak’s AI Adoption Maturity Model separates departments by stage, from experimenting with a single tool to embedding AI into core workflows, and maturity rarely moves in lockstep. Finance might pilot one use case while sales has already folded AI into daily work. Comparing departments against the same benchmark shows which teams are ready for a bigger investment and which need enablement first.

Where productivity signals reveal AI ROI

Adoption and maturity show where AI is used. Productivity signals show whether it’s working. Measuring AI’s actual impact means comparing output, focus time and utilization for the same team before and after adoption, not against generic industry benchmarks.

Signals that predict ROI

Three signals matter most: usage intensity, retention and output change. ActivTrak’s Productivity Lab found that employees who spend 7%-10% of their total work hours in AI tools have the highest productivity. A very small amount of employees actually fit into this category of power users; In fact, only 3% reach the maximum productivity level associated with an AI power user. Teams clustered below that range are strong candidates for enablement investment, while teams already above it may need governance rather than more tools.

Prioritizing AI spending across competing requests

Once you have adoption, maturity and productivity data for every team, competing requests stop being a popularity contest. Building a defensible AI roadmap means weighing each request against the same evidence instead of the size of the ask or the seniority of the requester.

Score requests using adoption, maturity and impact

Score each request across three dimensions: current adoption depth, maturity stage and the productivity signal it’s likely to move. IBM’s CEO research found an overwhelming majority of leaders (75%) note that “trusted AI is impossible” if an organization lacks governance, yet only 39% say they have good governance. This data serves as a reminder that low-maturity teams often need governance and enablement before more spend pays off.

Fund the team with the clearest data trail. Track the emerging AI ROI signals tied to each investment.

Make AI investment a data-driven, cross-team discipline

PwC’s Global AI Jobs Barometer found that industries most exposed to AI have seen productivity grow 34% in 2025 compared to 2018, proving that disciplined AI investment compounds over time.

Treat adoption data, maturity benchmarks and productivity signals as a standing scorecard. Every future AI request needs to be evaluated against the same evidence.

See where your next AI investment should actually go. ActivTrak’s work intelligence platform turns AI adoption data into a clear, defensible investment roadmap, showing you exactly which teams are using AI, how deeply and where that usage is translating into real productivity gains. Stop guessing and make smarter AI investment decisions.

FAQ

How do you decide where to invest in AI tools next?

Compare adoption depth, maturity stage and productivity impact across teams, then fund the teams with the strongest mix of unmet demand and demonstrated gains, rather than the team that asked first.

How do you build a data-backed case for AI spending?

Pair your own adoption and productivity data with third-party research on governance and impact, then show leadership the specific gap your investment closes and the signal you’ll use to measure it.

How do you avoid over-investing in AI tools that aren’t used?

Track usage retention and hours in tool, not license, counts. Idle licenses are common once a company runs several AI platforms, and cutting them frees budget for tools teams are actually adopting.

How do you prioritize AI spending across competing requests?

Score each request against the same three data points: adoption, maturity and productivity impact. Focusing on these three core points helps teams judge based on evidence.

Share this article

Meet the author

Array
Sarah Altemus
Principal of Performance & Transformation
Sarah Altemus is a Principal of Performance & Transformation at ActivTrak, where she contributes to the company’s research and advisory efforts focused on work intelligence in the AI era. Working with one of the world’s largest datasets on how work actually happ... Read more
Sarah Altemus is a Principal of Performance & Transformation at ActivTrak, where she contributes to the company’s research and advisory efforts focused on work intelligence in the AI era. Working with one of the world’s largest datasets on how work actually happens, she partners with global enterprises to benchmark performance, apply best practices and translate behavioral data into measurable improvements in productivity, workforce effectiveness and organizational design.

Sarah brings a decade of experience advising organizations through complex, large-scale transformations where workplace strategy, culture and business operations must evolve simultaneously. Her work spans global enterprises including Expedia Group, ExxonMobil and Wizards of the Coast, where she shaped the human-centered strategies required to sustain performance through periods of significant disruption — including headquarters relocations, mergers, operating model shifts and digital transformation.

At Expedia Group, Sarah directed change management for the relocation of 5,000 employees to a new headquarters, developing enterprise-wide readiness programs, behavioral research initiatives and cross-functional alignment strategies. When COVID-19 emerged during the transition, she supported the company’s pandemic response, enabling a rapid and coordinated shift to remote work at scale. At ExxonMobil, she supported leadership through the organizational and cultural complexities of one of the largest corporate headquarters projects in the world, alongside a concurrent merger integration.

Earlier in her career, Sarah advised enterprise organizations including Amazon, Nordstrom and Philips Healthcare on workplace strategy and new ways of working, applying human-centered research and design thinking to align employee experience with business performance. She also served as a researcher at APQC (the American Productivity and Quality Center), where she developed expertise in benchmarking, process improvement and organizational effectiveness.

At ActivTrak, she focuses on helping organizations operationalize work intelligence — enabling leaders to embed data-driven ways of working and drive adoption at scale. Her work emphasizes that sustainable performance gains require not just new technology, but a fundamental redesign of how work happens, supported by continuous measurement and organizational accountability.

Sarah’s areas of expertise include organizational design, workforce analytics, return-to-office strategy, employee listening at scale and change management in the context of AI and productivity technologies.
View author articles

Getting started is easy. Be up and running in minutes.