Useful intelligence does not start with competitors, but with the target customer

There is a stubborn bad habit in many companies: calling "competitive intelligence" what is really just a feature inventory. You look at what others have shipped, line up the new releases, tick the boxes, and then ask yourself what you should copy, catch up on, or launch in turn.

The problem is that this approach rarely produces any real strategic understanding. It tells you about what is visible, not about what is important. It shows you outputs, not necessarily outcomes. It looks at competitors, but not enough at the target customer.

And that is precisely where many teams go wrong.

Useful competitive intelligence does not start with competitors. It starts with the target customer. With their real needs, their decision criteria, their frustrations, their alternatives, their trade-offs, and what genuinely resonates with them. Only then do competitors become interesting: not for their own sake, but because they sometimes reveal something about how a market responds to those needs.

Jeff Bezos captures this reversal of perspective well:

"If we can keep our competitors focused on us while we stay focused on the customer, ultimately we'll turn out all right."

This point profoundly changes the way you do intelligence. Because the mistake is not only watching competitors too closely. The mistake is believing that the right question is: "what did they ship?" when the useful question is really: "what did they understand about the customer we're aiming for?" A feature is never anything more than a visible answer to an invisible line of reasoning. If you copy the answer without understanding the reasoning, you are not doing strategy. You are reproducing a gesture.

Output vs. outcome: an essential distinction

This is where a simple distinction becomes essential: the one between output and outcome.

The output is the feature, the interface, the module, the button, the AI agent, the integration. The outcome is the result obtained on the customer's side: time saved, a better decision, better coordination, less friction, faster learning, easier access to information. Poor intelligence stops at the output. Good intelligence tries to trace its way back to the outcome.

This difference also explains why copying a competitor is often a bad reflex. Two products can address the same need in very different ways. What matters is not producing the same thing, but producing an answer consistent with your positioning, your target customer, and your value proposition. On this point, April Dunford's work remains very useful: positioning does not just say what a product is, it says who it is the best choice for, against which alternatives, and for what value.

In other words, relevant competitive intelligence cannot be separated from positioning work. If you do not know exactly which customer you are aiming for, which alternatives you are fighting against, and why your product should be chosen, intelligence mechanically becomes superficial. It turns into a collection of misread signals.

Broadening the notion of a competitor

This also forces you to broaden the very notion of a competitor. A real competitor is not necessarily a product that belongs to the same category as you. It is anything that can resonate with the customer's need at the moment they make their choice. That is why tools like Notion are interesting to observe. Notion can encroach on project management, document management, tasks, the internal wiki, the knowledge base, and still other uses. It does not always "replace" a specialized product head-on, but it can be enough to prevent its adoption. This kind of alternative is sometimes more dangerous than a direct competitor, precisely because it is less visible on the usual radars.

The dimension of anticipation

Another frequent weakness of competitive intelligence is believing that everything that matters is already visible in the market. But that is not true. Not everything is in the competition. And not everything will be explicitly requested by customers either.

This is where the dimension of anticipation comes into play. Some important shifts first emerge in the form of weak signals: recurring frictions, workarounds, isolated but coherent requests, advanced uses among certain profiles, expectations observed in other tools. A company's role is not only to read the present. It is also to spot what could shape tomorrow's decision criteria.

Anticipating does not mean guessing the future. It means spotting weak signals, formulating hypotheses, testing early, and learning before others do. It is a logic of learning, not of divination.

And it is also a logic of risk. Because the point is never to choose between risk and the absence of risk. It is to choose which risk you accept. The risk of exploring too early, or the risk of arriving too late. What some companies call caution is sometimes just another way of letting others do the learning in their place.

The shift in expectation standards

This logic of anticipation is particularly important in a context where expectation standards are shifting fast. Customers do not compare your product only to your direct competitors. They also compare it, consciously or not, to the best software experiences they encounter elsewhere.

We have already seen this phenomenon with personal mobile apps. For a long time, many users accepted, in their professional environment, tools that were slow, inelegant, unintuitive, sometimes not real-time, on the grounds that they were "business" software. Then usage changed. Personal apps became more fluid, simpler, more responsive, more pleasant. And gradually, it became much harder to accept at work the tools we had considered normal a few years earlier. The standard for professional applications was raised by experiences that came from elsewhere.

AI is now amplifying this phenomenon at another level. It does not just change the design or the perceived fluidity. It changes the threshold of acceptability itself. Tomorrow, more and more users will find it hard to accept spending a long time searching for information a system could surface, producing manually what a piece of software could prepare, navigating complex interfaces to get a simple answer, or using tools incapable of helping, suggesting, summarizing, or speeding up the work. So the real issue is not just what direct competitors are doing. The real issue is the shift in how demanding customers have become.

This is where Kathy Sierra's quote takes on its full meaning:

"Upgrade your user, not your product. Don't build better cameras — build better photographers."

Customer-oriented competitive intelligence is not there to produce a list of features to imitate. It is there to better understand how to help a target customer succeed. This shift is decisive. It moves you from a catalog logic to a value logic.

Four levels not to confuse

In practice, this forces you to distinguish several levels. A benchmark compares visible elements. Intelligence seeks to understand movements. Positioning clarifies where your product can be the best choice. And strategy decides where you want to play, for whom, with what level of risk and what logic of differentiation.

Confusing these four levels is a classic source of error. You think you are doing strategy when you are stacking up observations. You think you are reading a market when you are only reading its surface.

Useful competitive intelligence is therefore less a monitoring activity than an interpretation activity. It does not aim to produce a map of competitors' features, but a better understanding of the target customer, of the alternatives that appeal to them, of the standards that are rising, of the weak signals that are emerging, and of answers consistent with your own positioning.

Copying your competitors is not a strategy. Understanding what they reveal about the customer, the alternatives, and the market's evolution, on the other hand, can become one. As long as you never forget that the competition is not the starting point. The target customer is.

The 5 questions of useful competitive intelligence

To keep competitive intelligence from being reduced to a simple feature benchmark, here are five questions that make it immediately more useful.

1. Which target customer am I observing?

Useful intelligence does not start from the competitor, but from the customer being aimed at. Before analyzing a market move, you need to know which type of customer it might make sense for, and why.

2. Which need or outcome is this competitor trying to address?

A feature is never the real issue. The real issue is the problem it is trying to solve, the gain it promises, or the friction it is trying to remove.

3. Is this a relevant move for our positioning?

The fact that a competitor is doing something does not yet mean we should do it. The right question is whether this move is consistent with our target customer, our promise, and our place in the market.

4. Which alternatives is this move trying to beat?

A competitor does not only fight against similar products. It also fights against habits, general-purpose tools, internal workarounds, or solutions deemed "good enough."

5. What does this signal reveal about the shift in market expectations?

Useful intelligence is not only there to read the present. It is also there to spot what is gradually becoming unacceptable, expected, or differentiating in customers' eyes.

Competitive intelligence starts to become strategic when it no longer produces a list of features, but a better understanding of the target customer, of the alternatives that really matter, and of the standards that are emerging.

To learn more

In software, the edge will no longer be technology. It will be understanding the context. The PM as Architect of Context