How we build safely at Birdie

Building care technology is a bit of a tightrope: you need to build products quickly, without compromising on safety. And as we know, you really can't afford to make mistakes here.

So how do you make sure you always get it right?

The answer is unfortunately not that sexy, but it works. It's process. At Birdie, our product development has five stages: idea, design, build, launch, and monitor. And safety isn't just a checkbox at the end, it's woven through every single stage.

In our latest "Under the Microscope" video from the Smarter Care Lab, Sarah (Product Manager at Birdie) walks through exactly how this works.

The tightrope: balancing speed and safety in care technology

In social care, people's wellbeing is at stake. The wrong feature design, a missed edge case, or inadequate testing can lead to real harm. A medication missed. A carer assigned when they're unavailable. Visit times changed without checking what's already scheduled.

But standing still isn't an option either. Care teams need better tools, faster workflows, and smarter technology to deliver quality care. The administrative burden is already too high.

This is the tightrope we walk every day at Birdie. And our four-stage safety process is how we stay balanced.

Our four-stage safety process explained

Stage 1: Early conversations

Whenever we're building something new, we sit down with Jenny, our Clinical Safety Officer, and ask a very important question: what could go wrong?

We work together to identify any risks by asking "what if…?" This happens with all kinds of features, such as visit times being changed without checking what's scheduled, or activities not being linked properly.

These early conversations force us to think through real-world scenarios before we're too far down the development path. We aim to identify risks before we've written a single line of code.

Stage 2: Testing with partners

Next, we test with our partners. This is what we call co-development: building solutions together with the people who'll actually use them. It's in our DNA at Birdie.

Why? Because you're the experts. You know how care actually works in someone's home at 7am when the carer's running late and managing multiple clients.

We show you early versions and ask: does this make sense? Could this cause confusion? Would your team actually use this? Your feedback shapes the solution and makes sure we've got every scenario covered.

Stage 3: Assessing the risks

Now we come to the formal assessment. We use something called SWIFT: that stands for Structured What-If Technique.

It works by rating risk through asking three key questions:

1. How likely is this to happen?
2. If it does happen, how serious or severe could it be?
3. Are there any existing controls in place?

The traffic light system: how we rate every risk

When we combine those factors, it tells us the risk priority—just like a traffic light.

Green: No or minor clinical risk

Some risks are green. Maybe it's a bit inconvenient, but there's no real harm. We can live with that.

Amber: Risk exists, but benefit outweighs it

Some are amber: there's risk, but the benefit outweighs it, and we should minimise the risk where practical. It's like crossing any road. The risk is there, but we accept it because we need to get to the other side.

Red: Unacceptable risks

And some risks are red: unacceptable. These must be fixed before we can release the feature. No exceptions.

Real-world example: medication warnings

Let's look at a specific example from our Upgraded Rostering product: medication scheduling.

We identified a potential risk: what if a medication isn't linked to a visit? Could someone miss their insulin? Could they get a double dose later because the carer didn't know it was already given? Time to put it through the traffic light system:

• Is it likely? It's possible. When you're scheduling 50 visits across multiple clients, it's easy to forget to link every medication.
• Is it severe? It potentially could be, particularly if it's an important medicine like insulin.

That combination puts it in the amber zone: there's risk, and we should minimise it.

Our solution? We've introduced warnings in Upgraded Rostering that flag when a medication isn't linked to a visit. The system alerts you immediately on the client profile, so you can't miss it.

Stage 4: Ongoing monitoring

But we don't stop there. Once a feature is live, we're watching. Is this risk happening more than we expected? Are there patterns we didn't predict?

We review insights that come in, track bugs, and analyse how features are being used in the real world. If we spot something, we go back. Maybe we thought a risk was acceptable, but it's happening more often than anticipated. That may change the risk rating, which may affect the priority of making improvements.

We review, we adjust, we improve. It's a constant loop.

That's how we aim to keep care safe at Birdie, considering safety at every stage, embedded into our product development process.

Stay in the loop with the Smarter Care Lab

At Birdie we're constantly developing new innovations to make social care smarter, safer, and more efficient. Want to stay updated on what we're building and how we're building it?

Visit the Smarter Care Lab to get the latest updates on our innovations and product developments.