When Does Data Become Insight?

Most mornings, I can find out quite a lot about myself before I’ve even got out of bed. I can see how long I slept and what my resting heart rate was overnight. By the end of the day, I’ll know how many steps I’ve taken and how active I’ve been. Add in connected scales, nutrition apps and everything else we can now track, and we can build a surprisingly detailed record of what our bodies are doing.

The strange thing is that I’m not sure all of that necessarily means I understand myself any better. If my weight changes tomorrow, is that significant or just a normal fluctuation? If I sleep badly tonight, does it matter? An individual number can tell me what happened, but it doesn’t tell me whether I should care. For that, I need to know what is normal for me, what happened before it and whether anything else changed at the same time.

This was one of the things that interested me when I spoke to Mark Talbot, Founder and CTO at Lifecycle Health. His interest in health data began with his own weight-loss journey. He had information coming from connected scales, an Apple Watch and nutrition tracking, but it sat across different places. Bringing it together allowed him to look beyond individual readings and see what was happening over time.

It sounds like a relatively simple distinction, but there’s a much bigger question underneath it. We’ve spent years finding more things to measure and increasingly sophisticated ways of measuring them. AI now gives us the ability to work through far more of that information than we realistically could ourselves. Perhaps the problem is beginning to change. Instead of asking what else we can measure, we might need to ask when all of that data actually becomes useful.

From Measurement To Meaning

Weight is a good example because one reading can be surprisingly misleading. Step on the scales one morning and discover you’re several pounds heavier, and it’s easy to assume something significant has changed. But that number could reflect something as ordinary as water. Put it alongside the days and weeks around it and you can start to see whether something has actually shifted.

The same applies to much of the information generated by wearables. A poor sleep score tells you something about last night. It becomes more useful when you know whether it was unusual for you and whether the change continued. You might also want to know what else was happening at around the same time. What matters may not be the measurement itself, but how it compares with your own normal and whether it forms part of a wider pattern.

That becomes more important as the amount of information available to us grows. Watches and other devices have become increasingly good at collecting data. Apps then turn it into graphs, scores and notifications. Yet every new measurement potentially creates another question about what we’re supposed to do with it.

At some point, collecting more information stops being the difficult part, understanding which bits deserve our attention becomes the harder one.

What Happens Between Appointments?

Most of us don’t see a doctor very often. When we do, we might be asked how we’ve been since our last appointment. It sounds like a simple question, but answering it can mean trying to remember several months at once. You might know you’ve been more tired recently or haven’t been sleeping as well, but remembering exactly when something changed is much harder.

During that time, a watch or other device may have been recording some of those changes every day. It might show how your sleep has changed, whether you’ve become more or less active or how your resting heart rate has moved over time. None of those measurements necessarily means very much on its own, but together they could add useful context to what you remember.

Of course, that doesn’t mean a doctor needs to see every step you’ve taken or every night’s sleep. Months of raw wearable data could quickly become more of a burden than a help. The challenge is finding the information that might actually matter.

This is something Mark has been exploring with Lifecycle Health. When speaking to doctors and specialists, he found that the problem wasn’t necessarily a lack of interest in the data. As he puts it, “they want the data; they’ve just never been able to get it.” What they need is a practical way to use it without having to work through thousands of individual readings.

Lifecycle Health brings information together within an individual specialist care practice, such as a women’s health clinic, so doctors can explore the data they already hold about their own patients. AI can then help them ask questions and look for patterns across that information. It isn’t there to diagnose someone or make a clinical decision. The doctor still does that.

That could change what happens when someone arrives for an appointment. Rather than starting with a broad “How have you been?”, the information collected in the weeks and months beforehand could help the doctor identify something worth focusing on. The patient’s own experience still matters, but the conversation can start with more context about what has been happening between appointments.

When More Data Isn’t The Answer

There is something slightly counterintuitive about all of this. We tend to assume that more information gives us a clearer picture, but Mark’s experience suggests that the amount of data isn’t necessarily the problem. His weight, activity and nutrition were already being tracked. What was missing was a way to bring that information together and understand what it was showing him.

AI makes that distinction more important. We now have tools capable of working through far more information than any of us could realistically analyse ourselves. That creates an obvious opportunity to find patterns across data that might otherwise sit unnoticed, particularly when information from different devices is brought together or builds up over months and years.

But being able to analyse more doesn’t automatically mean understanding more. AI can help surface a change or show that two things appear to be happening at the same time, but somebody still has to decide whether that relationship matters. In healthcare, that judgement belongs with the doctor rather than the technology. In other settings, the same basic principle applies: finding a pattern and understanding its significance are not necessarily the same thing.

Perhaps that’s where the opportunity becomes more interesting. We’ve spent years developing devices and systems that are exceptionally good at recording what happened. AI could help us do more with that information, not by giving us another measurement to look at, but by helping us see how the measurements we already have fit together.