Everyone’s talking about productivity like it’s a motivation issue.

Better habits. Better focus. Better routines. More grit.

But when you look at the UK’s digital capability data, a different story shows up.

Most people are not struggling because they are lazy or disorganised. They are struggling because the digital foundation inside organisations is often “good enough to function” but not strong enough to scale, standardise, or sustain high-output work.

And AI is not fixing that. If anything, it’s exposing it faster.

The surprising part: the data isn’t bleak

Here’s what I expected to see in the Lloyds Essential Digital Skills 2025 report: catastrophe.

What it actually shows is a country that is largely digitally functional.

  • 85% of UK adults have Foundation Digital Skills
  • 92% have Life Digital Skills
  • 82% have Work Digital Skills

That’s not doom and gloom. That’s “most people can get by”.

But “get by” is not the same as “work well”.

And this is the part leaders often miss. The issue is rarely total lack of skill. It’s fragile capability plus messy environments, which creates chronic productivity drag.

The report calls out one work task as the lowest-performing.

Using digital tools to improve personal or organisational productivity.

Only 68% of workers can do that, and it has declined year-on-year.

That stat matters because it tells you something uncomfortable:

Most workplaces are not leaking productivity because staff are incapable of their roles.

They are leaking productivity because staff are not enabled to work efficiently inside the tools, processes, and governance the organisation has (or hasn’t) put in place.

This is why tool adoption does not equal productivity improvement.

You can roll out Microsoft 365, a CRM, Asana, Notion, Slack, and five shiny AI tools, and still have a team that is drowning.

The hidden opportunity: “on the cusp” employees

Another stat that should be in every boardroom:

Only 45% of workers can complete all work digital tasks, but 28% are “on the cusp”, missing just 1 to 3 tasks.

This is huge.

It means a massive chunk of your workforce does not need a long training programme or a culture overhaul.

They need targeted closure of small gaps, paired with an environment that makes good behaviour easy.

This is also why generic training often fails. It treats everyone like a beginner.

Most people are not beginners. They are close, but blocked.

Blocked by messy file structures, unclear “source of truth”, undocumented processes, and tool sprawl.

The productivity tax no one sees

Here’s what this looks like in real life:

  • Searching for the “latest version” of a doc
  • Rebuilding something that already exists because nobody can find it
  • Copying and pasting data between systems
  • Waiting for the one person who “knows how it works”
  • Avoiding a system because it feels risky to touch
  • Using AI on top of a broken process, then wondering why the output is inconsistent

None of that shows up on a P&L line item.

But it absolutely shows up in delivery speed, margin, morale, and client experience.

This is why I don’t think productivity starts with a new tool, or even a new workflow.

It starts with digital capability and structural health.

Which leads to a simple, slightly spicy opinion:

Most productivity initiatives fail because they treat symptoms instead of the digital operating conditions that create them.

Why AI isn’t solving productivity

If AI were naturally driving productivity, we would expect digital productivity capability to rise.

It hasn’t. It has dipped.

That does not mean AI is useless. It means AI is an amplifier, not a foundation.

AI amplifies:

  • process clarity, if you have it
  • process chaos, if you don’t
  • governance, if it exists
  • risk, if it doesn’t
  • capability, if the team has it
  • confusion, if they don’t

So if your organisation is trying to “AI its way” into higher productivity without first stabilising the basics, you’re likely automating inconsistency and scaling risk.