Dera Blog
Analysis

Why So Many Companies Use AI but Still See No Real Productivity Gain

Adoption is high. ROI is much harder.

Rasin Ansar, Editor, Dera3 min readUpdated Jun 29, 2026
Analysis

Quick take

AI usage is widespread, but real productivity gains stay uneven when teams buy tools before they define workflows, owners, and success metrics.

Why trust this article

  • Written as a task-first editorial guide, not a paid ranking.
  • Connected to Dera task and tool pages for the practical next step.
  • Updated context is shown when the article record includes it.

Best next step

Use the linked task and tool pages below when you want a specific recommendation, workflow, or prompt instead of only commentary.

The AI adoption story is easy to tell. The productivity story is harder. Stanford’s 2026 AI Index points to broad organizational adoption, and McKinsey reports that most organizations now use AI in at least one function. But McKinsey also reports that only a minority see EBIT impact at the enterprise level.

That gap is the real story. Companies are buying access to AI faster than they are redesigning work around it.

Why the ROI gap happens

  • Teams use random tools without clear owners.
  • No one defines which tasks AI should improve first.
  • Workflows stay informal, so gains are impossible to measure.
  • Integration is weak, so people still do the same copying and cleanup by hand.
  • Leaders count logins or pilots instead of time saved or throughput improved.

Evidence from current research

McKinsey’s 2025 State of AI highlights the same pattern: AI is widespread, but financial impact is uneven, and workflow redesign is a key differentiator for high performers. Thomson Reuters’ 2026 reporting echoes the problem from another angle: the technology is moving into professional work, but organizations still struggle with execution, strategy, and operational readiness.

What actually works

  • Start with repeatable, high-friction tasks.
  • Assign a clear owner to each workflow.
  • Measure before and after with a real baseline.
  • Integrate the workflow into the systems people already use.
  • Use one strong recommendation per task instead of a random stack of tools.

How to measure real gains

A useful metric is cost or time per completed task. That could mean time to produce a sales brief, time to draft a customer response, time to ship a small code fix, or time to summarize a research packet. If the AI tool does not move a real operating metric, the pilot is probably theater.

Why Dera’s approach is different

Dera starts with the task. That sounds obvious, but it is the part many teams skip. The goal is not to say “use more AI.” The goal is to answer “what exact workflow are we trying to improve, and which tool is the best fit?” That is how productivity gains stop being abstract.

AI adoption is easy. Workflow change is hard. Productivity only appears after the second part.

Where to start

Pick one repeatable job this month. Define the input, output, owner, review standard, and target time saved. Then choose the AI tool that fits that one workflow well. That is how ROI becomes visible.

Sources and references

Related resources

The Dera Choice for this topic

Find the right tool

Go from the article to a tool, prompt, and route you can try.

Browse task guides