Blogs

What is a digital diet study? How to measure what people actually consume online

Research Methods
6
A digital diet study measures the content people actually consume across their feeds, item by item. Screen time tells you how long someone sat at the table. A digital diet tells you what was on the plate.

What is a digital diet study?

How to measure what people actually consume online, and why it isn't the same as screen time

Written by: Mark Nguyen, ND Research

Originally Published: November 21, 2026

Labels: Research Methods, AI-assisted, Continuously Maintained

Topics: Digital diet, Information diet, Screen time, Feed exposure intelligence, Consumer research methods, Digital wellbeing

TLDR: A digital diet study measures the content a person actually consumes across their apps and feeds, item by item. Screen time tells you how long someone sat at the table. A digital diet study tells you what was on the plate.

Key Ideas:

  1. A digital diet is the mix of content a person consumes online: the creators, topics, formats, ads and news that make up their feeds.
  2. Screen time and app usage data measure quantity (minutes, sessions). Digital diet studies measure composition (what was in those minutes).
  3. Because recommendation algorithms personalize feeds, two people with identical screen time can have completely different digital diets.
  4. At ND Research, digital diet studies use consented participants who share screenshots of their own signed-in feeds. AI classifies each item and links it to who the person is.
  5. The method is early. Exposure is not attention, samples are still small, and independent validation is ahead of us.

‍

Preamble

I first wrote down the idea of an "information diet" back in 2016, long before we had a way to measure one. During the pandemic, like a lot of people, our team spent far too much time scrolling, and we kept noticing how different each of our feeds had become. Nobody could really tell you what their own feed was made of, let alone compare it to anyone else's.

This piece defines what we mean by a digital diet study, how one works and what it can and can't tell you.

‍

What is a digital diet?

A digital diet is the overall mix of content a person consumes through their devices: which creators, which topics, which formats (short video, memes, long-form, news), how much advertising, and how all of that shifts over time. The food analogy is deliberate. Nutritionists care less about how long you spent eating than about what you ate.

The phrase "information diet" was popularized by Clay A. Johnson's 2012 book The Information Diet, which argued that people should be as deliberate about information as they are about food [1]. A digital diet study takes that idea from a personal habit to a research method: measuring diets across many people so you can compare groups.

‍

How is a digital diet different from screen time?

Screen time measures quantity. Phone settings, app analytics and passive meters can tell you someone spent 2 hours and 40 minutes on social media yesterday, and roughly which apps.

A digital diet measures composition. Think: two people both spend 90 minutes a day on the same short-video app. One's feed is mostly cooking creators, local news and hockey highlights. The other's is crypto commentary, gym content and luxury ads. Their screen time is identical. Their digital diets, and probably what they buy, believe and worry about, are not.

Both measures are useful. Screen time is good for questions about habit and dependence. Digital diets are good for questions about influence, culture and what a brand or message is actually competing against in someone's feed.

‍

How do you run a digital diet study?

Our process, in five steps:

  1. Recruit consented participants. We recruit paid participants through a programmatic research marketplace and a public participant site, whatsonmyfeed.org. People know exactly what they're sharing and why.
  2. Capture real feeds. Participants sign in to their own social accounts and upload screenshots of what their feeds are actually showing them, typically from two platforms. They can deselect anything personal before submitting.
  3. Classify every item. AI checks each screenshot is a real feed, then labels each post by platform, creator, content category (using the IAB Content Taxonomy), ad versus organic, sponsor brand and format [2].
  4. Link to the person. Each diet is connected to the participant's survey answers (demographics, brand use, attitudes) under a pseudonymous ID, so you can compare groups.
  5. Report and compare. For example: how do frequent buyers of a brand's feeds differ from the general population? Which creators show up again and again in a segment's feed?

Our internal feed-stability work suggests that a snapshot of around 20 posts per platform can capture much of what a feed tends to show a person, though that is an internal finding, not yet independently validated.

‍

What can a digital diet study tell you?

  • For brands: what your buyers' feeds actually look like, and which content sits around your ads.
  • For agencies and creator marketers: which creators are already present in a target audience's feed (a familiar voice in a familiar lane, rather than an interruption).
  • For researchers: how consumption differs by age, region, politics or identity, with real exposure data instead of recall.
  • For digital wellbeing: whether people who see their own diet change how they use their feeds. We're designing a 28-day randomized study with young Canadians on exactly this question, and plan to submit it for peer review.

‍

How does it compare to other methods?

  • Surveys and diaries rely on recall. People are poor at remembering what they scrolled past.
  • Screen time and passive metering measure time and apps, not the content inside them.
  • Social listening analyses what people publicly post, not what they're privately recommended.
  • Platform ad tools show how your own ads performed, not the rest of the feed.
  • Digital diet studies measure the content people were actually served, linked to who they are. They are smaller and slower than social listening, and more direct.

‍

What are the limits?

  • Exposure is not attention. Seeing what a feed served doesn't tell you what someone noticed. We're working on dwell-time measurement, but it's experimental.
  • Snapshots, not streams. Screenshots capture a moment. Continuous capture would be richer and more invasive.
  • Classifier accuracy. Our platform check performed at about 95% in early tests. Content labels have not yet been validated against a large human-labelled set.
  • Sample size. Treat small segments as directional.

‍

FAQ

What is the difference between a digital diet and an information diet?

We use them almost interchangeably. "Information diet" emphasizes news and ideas. "Digital diet" includes everything a feed serves: entertainment, creators, ads and memes.

Is a digital diet study the same as tracking someone's phone?

No. Participants choose what to share, from their own accounts, with informed consent, and can withdraw. We don't collect messages, passwords or contacts.

Which platforms can be studied?

Instagram, TikTok, YouTube, Facebook, Reddit and X.

Who runs digital diet studies?

ND Research runs them through its Tapestry platform, for brands, agencies and academic collaborators. Contact mark@nguyendigital.co.

‍

How to cite this

Nguyen, M. (2026, November 21). What is a digital diet study? How to measure what people actually consume online. ND Research. https://www.nguyendigital.co/blog-posts/what-is-a-digital-diet-study

‍

List of References

(Continuously Maintained)

[1] Johnson, C. A. (2012). The Information Diet: A Case for Conscious Consumption. O'Reilly Media.

[2] IAB Tech Lab. Content Taxonomy. https://iabtechlab.com/standards/content-taxonomy/

[3] ND Research. Nguyen, M. (2026, October 23). What is feed exposure intelligence? https://www.nguyendigital.co/blog-posts/what-is-feed-exposure-intelligence

[4] ND Research. "The internet is not equal (but not in the way you think)." (2026, February 17). https://www.nguyendigital.co/blog-posts/infodiet

Still curious?

Incomplete and outdated insights are a thing of the past. We give you a front seat into how your customers live, work and interact. We weave together diverse inputs into a complete picture of your product.