Operational efficiency: how I choose initiatives that actually matter

Operational efficiency is one of the easiest things to talk about but one of the hardest things to do well.

Most businesses I've worked with don’t lack ideas.

They lack focus.

If you ask any operations team what could be improved, you’ll get a long and accurate list:

• manual steps that shouldn’t exist
• old stuff that shouldn't be around anymore
• tools that don’t integrate properly
• processes that grew organically
• work that “just takes time”

The mistake I see the most isn’t ignoring those problems, it’s trying to address all of them at once.

That’s how you end up with activity, not impact.

 The danger of “1000 small improvements” (or death by 1000 paper-cuts)

I’ve worked through periods where teams were actively encouraged to raise improvement ideas (in fact I've encouraged them!) and they did. Backlogs filled up. Workshops were run. Boards were colourful. Progress was reported. Everything looked fantastic.

Six months later, very little had actually changed.

What happened?

The improvements were real, but they were disconnected. Everyone was busy improving their part of the system, while the system itself continued to behave exactly as it always had.

The result was:

• marginal gains spread thin
• fatigue from constant change
• and no clear story about what was actually better

“Lots of small improvements” feels healthy, but at scale it often creates noise. Efficiency doesn’t come from doing more improvement — it comes from choosing where improvement matters most.

Why I force myself to choose just two

These days, I deliberately try and limit myself to two operational efficiency initiatives at any given time.

Not because there aren’t more opportunities (there always are and always will be), but because focus is our most precious and finite resource.

Two initiatives forces hard decisions:

• what am I willing to leave alone for now?
• what pain am I explicitly accepting?
• where will effort actually move the needle?

This also sets expectations with teams. When everything is a priority, nothing is. When people know which two things genuinely matter, effort aligns surprisingly quickly and that is when you truely accelerate your outcomes.

How I decide what not to fix

This is the uncomfortable part for me.

I’ve had to learn to separate:

• things that are annoying
• from things that are expensive

Not expensive financially but expensive in time, attention, risk, or cognitive load.

There are problems I know exist that I will quite deliberately not address this quarter. Saying that out loud can feel so wrong, but pretending otherwise creates false hope and quiet frustration.

Choosing not to fix something is still a decision. Making it explicit builds trust.

The signals I look for

Instead of long improvement backlogs (don't get me wrong these absolutely do have a place), I watch for a small set of signals that consistently point to leverage.

1. Manual effort hotspots

If good people are regularly doing repetitive, low‑value work to keep things moving, that’s a signal.

In one environment, I saw highly skilled engineers were spending a surprising amount of time reconciling data between systems because “that’s just how it works.” No single source was crazy — but repeated hundreds of times, it also became costly.

That kind of manual task is a great candidate for an efficiency initiative. It doesn’t just save time — it frees attention for higher‑value decisions.

 2. Failure repetition

I don’t stress much about one‑off failures. I worry about patterns.

When the same type of issue shows up again and again (even if the underlying incidents look different) there’s usually a systemic cause:

• missing automation
• unclear ownership
• fragile handoffs
• or implicit knowledge that isn’t documented

These are high‑value fixes because they reduce both operational load and cognitive stress on teams.

3. Customer friction

Internal efficiency that makes life harder for customers isn’t efficiency, it’s just cost‑shifting.

Signals I pay attention to:

• repeated customer questions about the same thing
• workarounds customers shouldn’t need
• “we’ll raise a ticket for you” where self‑service should exist

In one case, the internal process worked fine. The customer experience didn’t. Fixing that single point reduced tickets, shortened resolution times, and improved trust, all without adding any new tooling.

 Measuring impact without vanity metrics

This is where I think a lot of good intentions fall apart.

I’m wary of metrics that look impressive but don’t change decisions. Dashboards that grow month by month usually signal misunderstanding, not maturity.

The measures I care about tend to be boring but very telling:

• time saved that actually returns to teams (not reallocated immediately)
• reduction in repeat work or re‑opened issues
• fewer escalations for the same class of problem
• qualitative feedback like “we don’t think about this anymore”

If people stop talking about a pain point, that’s often the strongest signal that something worked.

Efficiency is direction, not speed

The goal of operational efficiency isn’t to make people go faster. It’s to make the system work better.

When you choose a small number of initiatives deliberately:

• trade‑offs are visible
• progress is tangible
• and teams can see the connection between effort and outcome

That’s how efficiency compounds, not through endless improvement lists, but through disciplined focus.

Closing thought

Operational efficiency isn’t about fixing everything.

It’s about choosing where intentional effort will create disproportionate impact and having the discipline to leave other things alone for a while.

• Two initiatives.
• Clear signals.
• Honest trade‑offs.

That’s usually enough to move a system forward.

 Reduce noise.

Focus effort.

Operate with intent.