Nearly half of organisations lack a unified safety system. The largest companies? Some are running 200 different applications.
I'm not talking about small operations still figuring things out. These are large and established companies with proper budgets, compliance teams, and boards asking "why aren't we safer"? Clearly not because they don't have budget nor because they don't care. But because somewhere along the line, "collecting data" became the goal instead of "making better decisions". And this is never intentional.
Here's what I see all the time:
✅ Incident reports with pages of text that can't be compared, categorised, or analysed
✅ Data arriving whenever it feels like it, in whatever format makes sense to whoever's filling it out
✅ Leaders staring at dashboards that look busy but don't actually answer the question "are we getting safer?”
✅ Expensive software that is not fit for purpose and has been poorly adopted by teams.
I've lived this. You probably have too. Mountains of data that look impressive in a board presentation but can't guide a single decision.
The turning point isn't buying more sophisticated tools. I discovered that designing your data collection properly from the beginning is the answer.
When you build structure in tick boxes instead of essay questions, standard categories instead of "describe in your own words," scoring that means the same thing across every site, your data becomes useful. You can easily spot patterns, make comparisons, and actually answer questions. That’s the goal.
At our October webinar, we walked through exactly this with AI and data workflows. We showed the "before" systems and how AI and Automation made things more streamlined and actually saved lots of time. We also showed the exact ROI in terms of time and money saved.
→ Missed it?
Here is the REPLAY.
Join us Nov 13 for the next one: "Workflow, Not Wizardry. Register HERE.
Community Spotlight:
Rebuilding Fire Risk Assessment
Let me tell you about a retailer I once worked with.
They had fire risk assessments carried out across hundreds of sites. Every single one was filling out this massive form, pages and pages of open text boxes.
It looked thorough. It felt comprehensive. But it was almost completely useless. Why? Because there was no way to compare Site A's "medium risk" to Site B's "medium risk." No consistency. No categories. No ability to prioritise or track improvements over time. We couldn’t analyse for trends but of course, the bulky reports came in handy when it came to meeting legal requirements.
Here's what I did. I went back to the drawing board.
Removed repetitions: Yes some questions appeared more than once and some deserved to be combined. We ended up reducing the number of fields/questions by 65!
Added structure and built in scoring: Every answer from a choice-based question carried weight. No more guessing what "major" means.
Automated the outcome: One click, the system calculated a transparent, defensible risk rating. No longer based on “feeling”.
Created actions: Based on selected answers, corrective tasks were automatically generated with ability to assign owners and due dates.
The result? Fire risk assessments that took less time to complete, were easier to understand, and actually produced data we could use.
We took out subjectivity. No more "it depends". Instead, we had clear, consistent, actionable information.
The lesson? You don't need fancier or new technology. You need better-designed forms and processes, and you can start with what you have.
The Operator's Manual:
Audit One Form This Week
You don't have to overhaul your entire system.
Start with one form. The incident log, the hazard report, whatever your team uses most.
Look at it and ask yourself:
How much of this is open text boxes where people write whatever they want?
How much actually has structure like tick boxes, categories, standard options?
If someone asked me tomorrow "what patterns are we seeing"? could I answer them fast?
If the answer is "not really" or "I'd have to manually read through everything first," you've found your starting point.
Sometimes the fix isn't technology but rethinking the questions you're asking and how you're asking them.
🔥 What's Coming Next
You'll hear from me once a month with one thing you can actually use - could be AI-related, data, tech, digital tools or even a process (borrowed or H&S owned). Just practical approaches from people doing the work, using these tools and processes and seeing result.
November’s Webinar/Workshop: Workflow, Not Wizardry. How to borrow workflows from sales and marketing teams (trust me, they've figured out things we in safety haven't. Yet). Register here (riskandreturns.flocrunch.io)
December’s Quick Lab: Building a business case in 5 minutes flat.
Save the Date: VIGIL Summit 2026
October 26-30, 2026
This is a summit and global convening space that brings leaders, sponsors, and practitioners together. This is where safety leaders building evidence-led practices come together. Not another conference. A global shift in how organisations approach safety leadership, and technology.
Speaker nominations open next month. If someone's doing interesting work that deserves a spotlight, you should nominate them.
→ More on VIGIL and speaker nominations coming soon!
Resources (Email Exclusive)
A simple framework to evaluate whether your forms are actually helping or just creating work. Click the button below to get the checklist.
Technology isn't just about the latest gadget but about solving real-world problems that can genuinely improve people's lives.