Digitized Workflow Is Not the Same as Optimized Workflow

2026-09-23

Key Takeaways

• Digitization and optimization are fundamentally different concepts. Gartner’s own glossary defines digitization as converting a process from analog to digital form without any different-in-kind change to the process itself.
• As early as 1990, Michael Hammer argued in Harvard Business Review that most IT investment failed to deliver meaningful gains because companies used computers to run old processes faster, instead of redesigning the processes.
• Optimized workflow requires three things digitization alone doesn’t provide: redesigned process flow, carried-forward context, and a defined path for exceptions.
• A four-question framework can help pinpoint where a specific workflow sits on the digitization-to-optimization spectrum.

When a company says “we’ve digitized our workflow,” it usually means: an approval process that used to run on paper or email now runs in software, with electronic forms and digital signatures replacing physical ones.

That’s real progress. It’s also the source of a common misunderstanding: many operations leaders treat finishing digitization as finishing process optimization. In reality, it’s only the first step in a longer sequence.

This article separates two concepts that get conflated constantly — digitized workflow and optimized workflow — using both an industry-standard definition and long-standing evidence for why “computerizing” an old process rarely produces gains proportional to the investment.

→ Related: You Have Workflow. Why Is Work Still Moving Slowly?

What Digitizing Workflow Actually Achieves

Digitization solves exactly one problem, and only that problem: how information is stored and moved.

Gartner’s Glossary defines digitization as the process of converting a process from analog to digital form without a different-in-kind change to the process itself — in other words, digitization changes the medium, not the operating logic.

Applied to workflow, digitization delivers real value:

• Documents stop getting lost. A purchase request no longer sits forgotten on someone’s desk.
• History becomes searchable. Who approved what, and when, is recorded.
• Physical movement time disappears. No waiting for internal mail or a signature run.
• Raw data becomes available for later analysis — for instance, measuring the average time a request takes to get approved.

These are genuine, worthwhile improvements. The problem isn’t that digitization is useless — it’s that digitization solves only one layer of the problem, and many companies stop there, believing the job is done.

What Digitization Doesn’t Change

Digitization doesn’t automatically improve speed, decision quality, or a process’s ability to handle exceptions — because it doesn’t touch the process’s logic.

This is precisely the observation Michael Hammer made in his classic 1990 Harvard Business Review article, “Reengineering Work: Don’t Automate, Obliterate.” He argued that large IT investments were often disappointing because companies tended to computerize the old way of doing things — keeping the process intact and simply using computers to run it faster — instead of redesigning the process itself. That observation is more than three decades old, but it describes almost exactly what’s repeating today with the current wave of workflow digitization in many companies.

Specifically, digitization does not, by itself, change:

1. The number and order of steps. If a paper process had five sequential steps, the digitized version usually keeps the same five sequential steps — just running in software instead.
2. Who has authority to decide what. Old approval structures, even ones that no longer make sense, are typically carried over unchanged into the digital system.
3. How exceptions get handled. If exceptions used to be resolved by “calling the boss,” they still get resolved that way after digitization — there’s simply a system running in parallel that never gets used for those cases.
4. How much context travels with each step. An electronic form, like a paper one, usually contains only the fields it was designed to capture — it doesn’t automatically carry forward the reasoning, history, or related factors from the step before it.

If the underlying process had problems — too many unnecessary steps, misaligned authority, no real mechanism for exceptions — digitization will preserve and accelerate those exact problems rather than remove them. This is why so many companies that have “finished digitizing” still find the process just as slow and rigid as before.

What Optimized Workflow Requires

If digitization only changes the medium, then optimization (digitalization, in Gartner’s precise sense — using digital technology to change how a process actually operates) requires three additional layers:

1. Redesigning the flow, not just converting it. Before digitizing a process, the question to ask is: which steps genuinely need to be sequential, which can be combined or run in parallel, and which can be eliminated entirely if they no longer add real control value. This is what Hammer meant by “obliterate” rather than “automate” — not making the old way faster, but asking whether the old way still needs to exist.

2. Carrying context through every step. An optimized workflow doesn’t just pass along a form — it passes along the reasoning, related data, and prior decision history, so the person at a later step doesn’t have to ask again or decide with incomplete information.

3. Defining a clear path for exceptions. An optimized workflow specifies in advance what counts as an exception, who handles it, and ensures that exception is still recorded in the system — instead of falling outside it as an untracked phone call.

None of these three layers come from buying a better workflow tool. They come from re-examining the business process itself before deciding what technology will run it.

→ Related: Next-Generation Workflow: When AI and Organizational Knowledge Change How Work Operates

A Framework: Where Does Your Workflow Actually Sit?

Four questions can help pinpoint whether a specific workflow has stopped at digitization or has genuinely moved toward optimization:

1. Setting technology aside, has the underlying process actually changed compared to five years ago? If the honest answer is “no, only the tool changed” — that’s a sign of pure digitization.
2. Can the person at the final step explain why earlier steps decided what they did, without having to ask? If not, the workflow is missing context.
3. Was the last exception that occurred this month handled through the system, or through an informal channel? If it was informal, that’s a design gap — not an employee compliance issue.
4. If you measured average completion time for this process, did it drop significantly after digitization, or only slightly because paper stopped physically moving around? A small drop means most of the processing time still lives in the process logic, not the medium.

A workflow that answers “there’s a problem” to two or more of these questions is almost certainly still at the digitization stage, regardless of how modern the software running it happens to be.

Conclusion

The difference between digitization and optimization isn’t about technology — it’s about whether the process was redesigned or only converted to a different medium. A company can own excellent workflow software and still get the same result: slightly faster storage and retrieval, but no faster decision-making.

This is also why evaluating a workflow shouldn’t start with “which software should we use” — it should start with “does this process actually need to work the way it currently does.”

This article is part of a series on workflow, AI adoption, and operational management for manufacturing SMEs.

Related articles:

• You Have Workflow. Why Is Work Still Moving Slowly?
• From Approval Workflow to End-to-End Workflow
• Next-Generation Workflow: When AI and Organizational Knowledge Change How Work Operates

→ Complete the Digitalization Readiness Assessment