Theory

Aggregation Theory: Why Owning Demand Beats Owning Supply

Ben Thompson's Aggregation Theory: the three conditions, how aggregators commoditise suppliers, and why most companies calling themselves platforms are not aggregators.

Origin: Ben Thompson, Stratechery, 2015. Developed across subsequent essays into a body of work on aggregators, platforms and the internet's effect on value chains.
In short

Aggregation Theory is Ben Thompson's explanation of why internet-era companies gain power by owning the customer relationship rather than by owning supply. It holds that when distribution is free and transaction costs are near zero, the firm that owns demand can commoditise its suppliers, and that this produces a self-reinforcing cycle: more users attract more suppliers, which attracts more users.

When to use

As a lens for judging where power sits in a value chain you are entering, and whether your position accrues it or gives it away. It explains structure and does not tell you what to build.

What Aggregation Theory is

Aggregation Theory is Ben Thompson’s explanation of why internet-era companies accumulate power by owning the customer relationship rather than by owning supply. He set it out on Stratechery in 2015 and has extended it across a large body of subsequent writing.

The argument starts from a value chain with three parts: suppliers, distribution, customers. In the pre-internet economy, power sat with whoever controlled distribution, because distribution was expensive and scarce. The internet made it free, which did not merely reduce that advantage but relocated it entirely.

What is scarce now is attention and the customer relationship. A firm that owns those can set the terms on which suppliers participate, and suppliers become interchangeable from the user’s point of view.

  1. Three conditions, all required Most cases fail one

    Abundant supply, near-zero marginal cost, and a direct customer relationship.

  2. Distribution stopped being the lever The founding claim

    The internet made it free, which moved power to whoever owns demand.

  3. Aggregator and platform are opposites Not synonyms

    One commoditises its suppliers; the other strengthens them. Opposite strategies follow.

  4. It explains, it does not instruct And needs scale to apply

    Below the threshold you have the cost structure and none of the power.

The four conditions that have to hold at once before aggregation is available, and a map of the rest of this page. Miss any one and you have a marketplace.

The three conditions

Abundant supply

More supply than any user can evaluate, which makes discovery and curation valuable.

Every video ever uploaded. Every hotel room in a city. Every article published today.

Near-zero marginal cost

Serving one more customer costs essentially nothing in distribution or transaction.

A digital good delivered over a network you do not pay per unit to use.

Direct customer relationship

You own the user, rather than reaching them through somebody else.

The customer opens your app by name, and blames you when a supplier disappoints.

Ask who the customer blames when a supplier fails. If they blame you, you own the relationship. If they blame the supplier and think of you as where they found them, you are a directory.

All three are required and most claimed aggregators fail at least one. The third is the practical test, and it cuts through most of the self-description in this area.

The virtuous cycle

The mechanism is self-reinforcing once it starts, which is why aggregation produces such durable positions and why it is so hard to reach.

  1. Better discovery

    You solve the abundance problem better than the alternatives.

    Produces: Early users, attracted by curation rather than by supply

    Trap: Assuming supply attracts users. It is the other way round

  2. Users arrive

    The experience is good enough that people come directly.

    Produces: A direct relationship, unmediated by anyone else

    Trap: Arriving via someone else's channel, which is not ownership

  3. Suppliers follow

    Suppliers must be where the users are, so they accept your terms.

    Produces: Supply on your terms, at your margin

    Trap: Reaching for this before you have the users

  4. Suppliers commoditise

    From the user's side, suppliers become interchangeable.

    Produces: Falling acquisition cost and rising leverage

    Trap: Mistaking this for a moral failing rather than a structure

The cycle, drawn as a loop because it has no end point. Each turn makes the next easier, which is also why the position is close to unassailable once established and close to unreachable before it is.

Aggregator or platform

The distinction is the most useful practical thing in the theory, and the two words get used interchangeably by people who mean quite different structures.

An aggregator owns the customer relationship. Suppliers come on its terms and weaken over time, because users think of the aggregator rather than of who supplied the thing.

A platform lets suppliers reach customers while those suppliers keep their own relationship. The suppliers get stronger. Their brands survive, and they can leave.

These lead to opposite strategies. An aggregator invests in the user experience and in keeping suppliers substitutable. A platform invests in supplier tools and in making suppliers succeed. Building one while describing yourself as the other is a reliable way to spend money on the wrong things.

When to use it

Run it when
  • You are entering a value chain and want to know where power currently sits.
  • You depend on a large intermediary for customer access.
  • You are deciding whether to own the customer relationship or to serve people who do.
  • A competitor is described as a platform and you cannot tell whether that is accurate.
  • You are choosing between marketplace and direct models.
Do not run it when
When the lens is informative, and when it is the wrong one. It reads a value chain; it does not tell you what to build.

Aggregation in practice: Netflix and Blockbuster

One firm owned the distribution. The other owned the relationship with demand. Only one of those turned out to be the durable asset.

Case study It worked

Netflix and Blockbuster · 2000 to 2010

Blockbuster could see exactly what Netflix was doing, and could not follow without shooting itself.

Blockbuster made a substantial share of its profit from late fees. Netflix, by design, had none: a flat subscription, no due dates, and a queue. That was not a feature Blockbuster had failed to think of. It was a feature Blockbuster could not copy without deleting a revenue line that funded the stores.

Blockbuster did eventually respond, dropping late fees in 2005 at a reported cost of around $200m a year, and launching Total Access. It filed for bankruptcy in 2010.

The frequently told version of this story is that the incumbent was blind. It was not. It was correct about its own economics, and the correctness is what trapped it.

Reported cost of dropping late fees
~$200m per year
Blockbuster stores at peak
~9,000
Outcome
Chapter 11, 2010

What it shows: This is counter-positioning, and the test is not whether the incumbent notices. It is whether the incumbent, having noticed, is better off doing nothing. Blockbuster was, for years.

Source: Hamilton Helmer, 7 Powers, 2016; Blockbuster SEC filings and Chapter 11 petition, 2010.

When it won’t help you

  • It describes a small number of very large companies

    The theory was built to explain Google, Facebook, Netflix and Amazon. The three conditions are demanding, and most businesses fail at least one, usually the abundance of supply or the direct customer relationship.

    Instead: Use it to understand the chain you sit in, not as a description of what you are building. Being a supplier to an aggregator is an ordinary position, and knowing it is one is useful.

  • It is a theory of outcomes, not of how to get there

    Aggregation explains where power settles once the cycle is running. It has nothing to say about reaching the threshold, which is the entire problem for anyone who is not already there.

    Instead: Treat the endpoint as a description and get your route from something else.

  • It has been criticised as unfalsifiable in places

    Because aggregation is identified after the fact, almost any successful internet company can be fitted to it retrospectively, and the boundary between aggregator, platform and marketplace is drawn differently by different writers.

    Instead: Hold to the three conditions strictly. Anything that fails one is something else, whatever it is called.

  • Regulation now shapes the outcome as much as the mechanism does

    The theory largely predates the current wave of competition and platform regulation, which directly targets the supplier-commoditising behaviour it describes.

    Instead: Read the mechanism as one force among several rather than as a prediction about how a market will settle.

Four honest limits. The first rules it out for the large majority of companies, including almost every early-stage one.

Further reading

  • Ben Thompson, Aggregation Theory, Stratechery (2015). The original essay, and still the clearest statement.
  • Thompson’s subsequent Stratechery writing on aggregators and platforms, where the distinction is worked out in detail.
  • 7 Powers. Network economies, which is aggregation expressed as a barrier.
  • Disruptive Innovation. The other main explanation of how incumbents lose.
  • Playing to Win. Where in the chain you are choosing to compete.

How to apply Aggregation Theory

  1. 1

    Map the value chain into suppliers, distribution and customers

    Who makes the thing, who gets it to people, and who consumes it. The theory's claim is that the internet removed distribution as a source of power, so whoever previously controlled it has already lost, whether or not they have noticed.

  2. 2

    Check whether supply is genuinely abundant

    Aggregation requires more supply than any user can evaluate. Where supply is scarce or physically constrained, suppliers keep leverage and the mechanism does not run. This is the condition most misapplications skip.

  3. 3

    Check your marginal cost of serving one more user

    Aggregation needs near-zero marginal cost for distribution and transaction. If serving each additional customer costs you real money, growth does not compound into power, it just costs more.

  4. 4

    Ask who owns the relationship when something goes wrong

    The clearest practical test. If the customer blames you when a supplier fails, you own the relationship. If they blame the supplier and you were merely where they found them, you are a marketplace or a directory, not an aggregator.

  5. 5

    Work out whether your suppliers get commoditised or strengthened

    On an aggregator, suppliers become interchangeable from the user's side and their brands weaken over time. On a platform they keep their own customer relationship and get stronger. These lead to opposite strategies, so knowing which you are building matters early.

  6. 6

    Be honest about whether you can reach the threshold

    The virtuous cycle only starts once you have enough users that suppliers must come on your terms. Below that threshold you have the cost structure of an aggregator and none of the power, which is an expensive place to sit.

Common mistakes

  • **Using aggregator and platform interchangeably.** They are opposite structures. A platform's suppliers keep their customer relationship and get stronger; an aggregator's lose it and get commoditised.
  • **Assuming any marketplace is an aggregator.** If users blame the supplier rather than you when something goes wrong, you are where they found each other, not the owner of the relationship.
  • **Applying it where supply is scarce.** The mechanism needs abundant supply. Where supply is physically constrained, suppliers keep leverage and no amount of demand aggregation removes it.
  • **Ignoring the marginal cost condition.** Zero marginal distribution cost is a precondition, not a detail. With real per-customer costs, scale increases spend rather than power.
  • **Aspiring to aggregation from a subscale position.** Below the threshold you carry an aggregator's cost structure with none of the leverage, and that is a difficult position to fund your way out of.
  • **Treating it as a strategy.** It is an explanatory theory about where power accumulates. It does not tell you what to build or how to reach the threshold.

How ShipFit operationalizes this

ShipFit runs Aggregation Theory in Stage 4 (How to Win?), alongside Platform vs Pipeline, to establish where in the value chain a proposed business sits and whether that position accrues power or gives it away. For most early-stage software the honest answer is neither aggregator nor platform, and the stage's follow-up is which of the [7 Powers](/frameworks/7-powers) is actually available at your phase.

Part of a larger playbook

ShipFit runs 55 frameworks across 9 decision stages

Aggregation Theory is one tool in a bigger toolkit. The full library covers market sizing, buyer discovery, MVP scoping, pricing, and launch.

shipfit.ai/frameworks
Frameworks Library
55 frameworks, mapped to 9 stages

The Mom Test

Q3

Rob Fitzpatrick

Validation question methodology, real interviews, not theater

Jobs-to-be-Done

Q2-Q4

Clayton Christensen

Functional, social, and emotional jobs your product fulfills

7 Powers

Q4

Hamilton Helmer

Strategic moats: Scale, Network, Counter-positioning, Switching, Brand, Cornered Resource, Process

Van Westendorp PSM

Q6

Feature-weighted price sensitivity analysis without guessing

Blue Ocean Strategy

Q4

Kim & Mauborgne

ERRC framework: Eliminate, Reduce, Raise, Create

Fake Door Testing

Q7

Pre-build behavioral validation with landing pages and apology modals

+ 49 more: TAM/SAM/SOM Analysis, Porter's Five Forces, Market Timing Analysis, Unit Economics (LTV/CAC)...

Frequently asked questions

What is Aggregation Theory?
Ben Thompson's explanation, first published on Stratechery in 2015, of why internet-era companies gain power by owning the customer relationship rather than by owning supply. Its argument is that the internet made distribution free and transaction costs near zero, which removed the traditional source of leverage in a value chain. The firm that owns demand can then set the terms on which suppliers participate, and suppliers become interchangeable from the user's point of view.
What is the difference between an aggregator and a platform?
Who owns the customer relationship. An aggregator owns it, so suppliers come on its terms and become commoditised over time: users think of the aggregator rather than of who supplied the thing. A platform enables suppliers to reach customers while the suppliers keep their own relationship, so they get stronger rather than weaker. The words are used interchangeably in casual writing and describe opposite structures with opposite strategies.
What are the conditions for aggregation?
Three. Supply must be abundant enough that no user can evaluate all of it, which is what makes discovery and curation valuable. Distribution and transaction costs must be near zero, so serving one more user costs essentially nothing. And you must own the customer relationship directly rather than reaching customers through someone else. Miss any of the three and the mechanism does not run, which is why it applies far less widely than the term's popularity suggests.
How do I know if I own the demand relationship?
Ask who the customer blames when a supplier fails. If they blame you, you own the relationship and the supplier is interchangeable from their side. If they blame the supplier and regard you as where they happened to find them, you are a directory or a marketplace. It is a practical test and it cuts through most of the self-description in this area.
Can a startup become an aggregator?
Rarely, and aiming at it early is expensive. The virtuous cycle only begins once you have enough users that suppliers must accept your terms, and below that threshold you carry an aggregator's cost structure with none of its leverage. In 7 Powers terms, network economies are a takeoff-phase power rather than one available at origination, so a new entrant reaching for aggregation is usually reaching past what its phase makes possible.
Related on ShipFit

Keep exploring

Master guide
Validate your business idea

The 9-step playbook from market verdict to ship-ready spec.

Framework
7 Powers

The 7 Powers, each with its benefit and its named barrier, plus the Power Progression that decides which of them a startup can realistically build and when.

Framework
The Lean Startup

Validated learning, the build-measure-learn loop, what an MVP actually is, the three engines of growth, and the ten pivots Ries names rather than one.

Guide
Market Research

Most founder market research is a TAM slide that nobody believes. The numbers that actually matter are smaller, harder to defend, and tell you whether the market exists for the ten-customer version of your business.

Guide
Idea Validation

Most founders confuse idea validation with idea-receiving-encouragement. The two have nothing in common. Here's what real validation looks like, and the four methods that actually produce it.

Calculator
Startup valuation calculator

What's your SaaS worth? Low, mid, high, in three inputs.

Q&A
How do I find my target market?

Narrow in four passes. (1) Start with the broad category your idea sits in. (2) Filter by buyer behavior: who currently has this problem and is doing something about it? (3) Filter by reach: who can you actually contact via the channels you have today? (4) Filter by willingness to pay: who has budget authority and a price point that clears your unit economics? The output is a specific buyer profile you could name 10 people who match. If you can't, you haven't narrowed enough.

For founders
SaaS founders

SaaS idea validation that pressure-tests your ICP, pricing model, and retention before you build. ShipFit forces 9 decisions in ~20 minutes. Start free.

Comparison
Buildpad

If you want a conversation partner, Buildpad. If you want to stop researching and ship, ShipFit. Both solve different problems for different founders. Don't pick on hype.

Ready to make your next product a success?

9 decisions between your idea and a product worth building.

No credit card required.

Try an example: