Andrew Wiseman looks at the latest updates to the UK’s inflation basket, and asks whether the dataset is useful, underused, or both.
Every year, the Office for National Statistics quietly updates the UK’s inflation basket. Every year, almost nobody outside economics pays much attention.
Which is a shame. Because if you’re interested in how people actually live, think and behave, it might be one of the richest and most underused datasets we have.
This year’s update is a good example. Houmous is in. Alcohol-free beer is in. Pet grooming, dashboard cameras and motorhomes are in. Premium lager in pubs is out. Sheets of wrapping paper are out. On the surface, it is a technical adjustment. The basket exists to measure inflation. Items are added and removed to reflect what people are actually buying. But step back for a moment and it starts to look less like an economic tool and more like a cultural artefact.
The most boring dataset in Britain?
There is a tendency in research to privilege the shiny things. Social listening. AI. TikTok trends. Neuromarketing.
Meanwhile, sitting quietly in the background, is a dataset that is nationally representative, behaviour-based, updated annually and rooted in actual spend.
The inflation basket is a lagging but highly reliable map of mainstream behaviour. Not what people say they do. Not what they aspire to do. But what has crossed the threshold into normality. That distinction matters. Because in research, the most interesting space is often not what is new, but what is no longer new.
When a trend stops being a trend
Take alcohol-free beer. For years, it has been a trend. A talking point. Something highly visible in media and often over-claimed in surveys. Now it sits in the basket.
Which tells us something quite specific. It has moved from emerging to embedded. From niche to normal.
The same is true of houmous. At one point a slightly middle-class curiosity. Now a staple.
This is where the inflation basket becomes useful for researchers. It does not tell you what is about to happen. It tells you what has already happened, but perhaps has not yet been fully recognised.
That lag has value. It helps separate what people are talking about, what they say they are doing, and what they are actually doing consistently. Most research sits somewhere between the first two. The inflation basket sits firmly in the third.
The story behind the story
Look at the additions this year and a pattern starts to emerge.
- Health signalling shows up in alcohol-free beer and houmous.
- Risk awareness appears in dashboard cameras.
- Emotional spending is visible in pet grooming.
- Lifestyle aspiration sits behind motorhomes.
None of these are surprising on their own. Taken together, they point to a consumer who is more deliberate in how they spend. There is evidence of restraint, but also of selective indulgence. Spending has not disappeared, it has been redirected towards things that feel justified or meaningful.
And the basket captures this through behaviour rather than explanation.
Behaviour is not explanation
There is a useful tension here. If you ask people why they are buying alcohol-free beer, you will get rational, socially acceptable answers. Health. moderation. better choices.
All valid. None complete. The basket does not attempt to explain. It simply reflects that enough people are buying something, often enough, for it to be considered representative.
Which raises a question. How often do we treat people’s explanations of their behaviour as if they are the behaviour itself? And how often do we build strategies on that assumption?
From observing behaviour to interrogating it
If datasets like the inflation basket show us what has become normal, the next question is what we do with that. Behavioural data can feel complete. It shows what people buy, how often, and in what combinations. But it does not tell us what that behaviour means, what sits behind it, or what might cause it to change.
That is where research still has a clear role.
A different kind of role for research
The growth of passive data has changed the landscape. We no longer need to ask people whether they buy alcohol-free beer or how often they eat houmous. We have better sources for that.
The role of research shifts away from measurement and towards interpretation. It sits alongside datasets like this and focuses on the questions those datasets cannot answer.
Questions about meaning, context and tension.
Moving up a level
The real opportunity is to treat datasets like the inflation basket as a starting point. Noticing that alcohol-free beer has entered the mainstream is useful. Understanding why this category has crossed the line, why now, and where it still does not quite fit is more valuable.
Those are the questions that inform decisions.
Where this approach pays off
This way of working becomes particularly useful in a few areas.
- Understanding what is genuinely established. When something appears in a dataset like this, it has moved beyond early adoption. The conversation changes.
- Exploring trade-offs. Behavioural data shows where money is going. It does not show where it has been taken from, or how people justify those choices.
- Finding the edges. Every behaviour has limits. There are situations where it does not work, or groups for whom it does not resonate. Those edges often point to opportunity.
A more useful division of labour
It helps to think about this as a division of labour. Behavioural datasets show what is happening at scale. Research helps us understand what that behaviour means.
Each on its own is partial. Together they are more useful.
- If you want to understand where culture might be heading, there are plenty of places to look.
- If you want to understand what people think they are doing, you can ask them.
- If you want to understand what Britain actually is, in a quieter and more habitual sense, the inflation basket is not a bad place to start.
It is an unassuming dataset. Which is precisely what makes it worth paying attention to.