#126 - Three reports, but only one truth!

8 Oct 26

Every week I produced the same information three times, in three different formats, and for three different recipients. It cost me two hours a week. Across my peers, it cost the company around 40. Nobody was asking whether any of it changed a decision.

Trial Brief

Data drives decisions, and reporting is how that data reaches the people who make them. That makes reporting a necessary part of running engineering projects. The overhead of doing it badly is rarely counted. When the same information has to be reshaped for every recipient, the cost isn't just the hours spent building powerpoint slides, it's the engineering time quietly taken away from doing actual engineering!

Trial Execution

I was once seconded to a major commercial entity where I found myself reporting the same information in three different powerpoint formats every week, for three different recipients. Some wanted traffic lights, others wanted arrows showing whether a trend was up, down or static. The underlying data was identical, only the packaging differed.

Producing those reports took around two hours of my time every week. Multiplied across my peers, the company was losing roughly 40 hours a week to inefficient reporting. I raised it as an issue, and nothing was done.

After six weeks of waiting, I decided to act unilaterally. I created a single common format designed to satisfy all three recipients and started sending it out. The pushback was significant and immediate. They also told me they 'got the issue' and would address it. Of course, they never did.

Trial Findings

The tooling isn't the hard part any more. Data analytics are far more advanced than they were, and software suites such as Tableau make it technically straightforward to generate tailored reports from a single source of data. But moving an organisation from hand-built slides to automated, tailorable reporting still takes time. The time it takes is mostly human, not technical.

It requires deft stakeholder management. Someone has to get agreement on what should be measured, who should receive it, and how often. Each of those is a negotiation, because every recipient has built their own picture of what 'good' looks like, and changing a format can feel like losing visibility or control. That's likely part of why my common format drew such an immediate reaction, and why 'we get the issue' so often stops at acknowledgement.

Underneath it all sits a simple principle: data drives decisions. If the right data doesn't reach the right people at the right time, decisions get made on instinct, anecdote or whoever spoke last. Former Hewlett-Packard CEO Lew Platt famously said: "If only HP knew what HP knows, we would be three times more productive." The knowledge existed inside the organisation. It simply wasn't surfacing where it could be used. That's why knowing what to measure is so important, and why it can't be left to the whim of personalities. If the metrics in a report depend on who happens to be in the role, or who shouts loudest, the organisation ends up measuring what individuals prefer rather than what the business needs. What to measure should be agreed deliberately, tied to the decisions it supports, and owned independently of any one person's preferences.

The most important question sits underneath all of that: what information do people actually need, as opposed to what they want, and what will they do with it? A report that doesn't inform a decision or trigger an action is overhead. A beautifully formatted report produced as a vanity project, to satisfy someone's ego rather than to support the business, is waste dressed up as diligence.

Applying that test is uncomfortable, because it forces a conversation most people would rather avoid: if nobody can say what decision a report supports, why are we even producing it in the first place?

The Common Thread

Data drives decisions, so reporting should serve decisions, not habits, and what gets measured should be a deliberate choice rather than a product of personalities. The cost of poor reporting is paid in engineering hours that never show up on a balance sheet, multiplied across every person doing the same thing slightly differently. Better tools help, but the real fix starts earlier, with an honest agreement about what needs to be measured, who needs it, how often, and what they'll do with it.

Trial Sign-off

One practical action

Pick one recurring report you produce or receive. Ask the recipient one question this week: 'What decision does this help you make?' If the answer is vague, you've found a candidate to simplify, merge or stop.

One question for reflection

Where in your organisation is the same information being re-packaged for different audiences, and who owns the job of getting those audiences to agree on one version?

Have a great week!

Steve

If you're interested in through-life Engineering Leadership development from first time Engineering leader/manager, through Head of Engineering up to Director, see the link here.

For short courses which target specific leadership issues, see this link.