A crucial, yet risky, phase in creating a new product is product discovery. It is the initial process of researching and validating customer needs and market opportunities to decide what to build and ensure it delivers real value to users. Execute it well to dramatically increase chances of product success. Do it poorly to increase the chances of wasting resources on a product nobody wants.
Product teams have often two different extreme approaches to product discovery:
- Skipping it all together: “if we build it, customers will come”.
- Overanalyzing it: “let’s dive deep and research everything before we move forward”.
The first path risks wasting time and money on the wrong product. The second can lead to endless analysis that delays—or even prevents—launch. The challenge lies in finding the balance between speed and accuracy.
And that balance often comes down to decision-making. Since there is no universal playbook for product discovery, teams either rush into design without validating the value proposition, or they adopt heavy frameworks that slow progress. That is why good decision-making during product discovery is the true job of any product leader.
When (over-)analysis falls short
As an engineer by training, I’ve always been a strong advocate of structured, analytical decision-making. My guiding principle has long been: “the bigger the impact, the more information we need before moving forward”. In fact, during my initial consulting years working for multinational companies, much of my work, as part of a team, focused on defining processes and interpreting decision frameworks, so that leaders could act with confidence, almost on “auto-pilot“. When we clearly explained the process and engaged senior management early, decisions were well-received and outcomes accepted. Even when a decision turned to be wrong, it was usually justified by unexpected events or factors beyond anyone’s control.
But when I began working as a product manager in small and mid-sized companies, things changed. I tried to apply the same structured and analytical decision-making processes to prioritize features or to define and choose market segments. But the process took too long and stakeholders soon started to become impatient. They did not care much about the decision making process. They just wanted the product to move in the right direction, and quickly. Somehow it all felt like being in a race. It became clear that I needed a more efficient, yet still reliable, way of making decisions. That’s when, after some research, I came across “heuristics”. Soon I realized how powerful they can be for making fast, effective choices during product discovery.
Moreover, I found out that analytical methods work best for products in existing markets. In these cases, deciding what to build or which features to prioritize is easier. Market data is available and customers can clearly express their needs. On the other hand, when a product targets a new or re-segmented market, competitive data is inexistent or irrelevant, and potential customers are hard to find. This often translates into greater uncertainty.
Heuristics in product discovery
Before looking at how heuristics apply to product discovery, let’s first clarify what heuristics actually are.
Gerd Gigerenzer, the German scientist known for his work on heuristics, argues that analysis works best in hindsight: explaining what has already happened. But, in complex and uncertain situations, analytical methods often fall short, while heuristics can be highly effective. Far from being irrational, heuristics are efficient mental shortcuts that save time and effort by ignoring some information to make decisions faster and with enough accuracy. The term itself comes from the Greek word meaning “to discover,” reflecting how heuristics rely on analogy, generalization, or other inductive processes to guide action when perfect solutions are out of reach. Some people refer to heuristics as “rules of thumb”.
In product discovery, the main goal is to repeatedly come up with new hypotheses, validate or disprove them. Hence, speed and clarity in decision-making are very important to move on. Using the right heuristic helps interpret hypotheses and observations through simple, practical frames, leading to “good enough” judgment. However, not all heuristics work equally well in a specific context. The right choice can accelerate progress, while the wrong one can lead to failure.
So, here are the steps I follow to apply heuristics in product discovery.
Step 1 – Contextualize the decision
The first step in using heuristics during product discovery is to set the context. Contextualizing means highlighting some aspects of a situation while downplaying others. Too often, we underestimate this step. Yet context is critical. An approach that works in one setting may fail in another.
Context can refer to many factors: the target market, product vision, company principles, competition, distribution channels, user expectations, or technical feasibility.
For example, imagine we are deciding whether to launch a product in a new market. We have already identified the target market segment. The context of our decision is now: “validation of the needs of the market segment”.
Step 2 – Formulate the hypothesis
Heuristics are no excuse for shortcutting the process of formulating hypotheses and gathering observations. Without a hypothesis, there would be nothing to decide about. So, returning to our example, if our context is “validating the needs of a market segment for a premium food delivery service”, we may need two hypotheses to test:
- Hypothesis 1: Single, health-conscious office workers in Milan don’t cook meals but order food online or eat out at least three times a week.
- Hypothesis 2: This same group is dissatisfied with the nutritional quality of available food options when eating out or ordering online.
Step 3 – Choose a heuristic
Once we have hypotheses, the next step is choosing the right heuristic. The choice depends on the domain. So, talking to domain experts helps us confirm which heuristics fit and even discover new ones. For example, in usability research, instead of running costly and complex experiments with many user types, Jakob Nielsen recommends using 10 simple heuristics to guide interface design.
I personally have a set of heuristics that I have used in different product discovery phases and still rely on today. They are further below in this post.
It is important to choose a heuristic before gathering observations. This is not easy to apply in reality but our mind can get easily biased from data, pick up the wrong information, and influence our choice for a proper heuristic. Conversely, first we should choose a heuristic and then we should go after relevant observations to take a decision.
Step 4 – Gather enough observations
After selecting a heuristic, we still need observations before making a decision. Without them, the risk would be too high.
Observations give us input from the real world. They can come from surveys, web searches, research papers, or even personal experience. Anything that informs our decision is useful.
Analytical methods usually demand large amounts of reliable data. Heuristics, instead, work with just enough input to keep the process lean. Sometimes a quick and focused web search is enough. Other times, a couple of customer interviews will do. The goal is accuracy with speed—we gather enough evidence to feel confident, then move forward.
It helps being explicit about the type and number of observations we need before we start collecting them. This prevents us from cutting the process short out of convenience or fatigue. In customer needs analysis, statistical significance is often seen as the holy grail. But in practice, reaching that many responses can be costly and hard, especially within a niche target segment. A more pragmatic approach is to decide upfront how many responses will give us a reasonable level of confidence. This way, we stay efficient—guided by common sense and informed intuition.
Step 5 – Decide
Decisions can be positive, negative, or about making changes and iterating. Sometimes, new observations show that our initial hypothesis was off. In that case, we adjust it and gather fresh input before reaching a conclusion. So, the whole process can be iterative. On the other hand, with the right heuristics, it may take only seconds to decide. That is the good side of heuristics: little effort for maximum impact.
6 heuristics for product discovery
Here are some of the heuristics I use in product discovery. The list isn’t exhaustive, but it shows what heuristics can look like in practice. Each of us can develop our own set, shaped by our industry, product type and personal experience.
| Domain | Heuristic | Product Discovery Application |
| Competition | Recognition | If customers in our target segment spontaneously name a brand, product, or alternative solution to her need, we’ve identified our top competitor or reference solution. |
| Customer Feedback | Rule of Three | If at least three different customers independently raise the same issue or request, we treat it as a validated signal worth exploring. |
| Needs Analysis | Information Availability | When customer data is limited, we use the insights available rather than waiting for statistical significance. For example, if more than 30% of interviewed customers (at least 2 out of 5) mention the same need or pain, it’s worth addressing. If the number rises to 60% (at least 3 out of 5), the need becomes critical to solve. |
| Product Vision | Don’t Break Ranks | Any feature or process that doesn’t align with our company’s strategy, mission, and values is irrelevant. It can even be harmful. |
| Prioritization | Default | As the Kano model suggests, we prioritize features that users give for granted. Then we can add delight features, that excite when present but don’t frustrate when absent. |
| Feature Selection | Fast & Frugal Tree | First, we drop features that require excessive effort. From the rest, we drop low-impact features (those not solving a clear pain point). Finally, we discard features unlikely to make a difference in the next few months. |
Wrapping up on smart decisions in product discovery
The greater the impact of a decision, the more information we typically need to minimize the risk of failure. Yet, when uncertainty is high, analytical decision-making can become time-consuming without producing meaningful insights. In these situations, heuristics (simple rules of thumb based on experience and expertise) often provide a more effective way forward.
That said, analytical tools still play an important role. Even in uncertain environments, they help teams explore the context, structure information and stimulate collaboration. Sometimes, the real value of analytical tools often lies in the process itself, more than in the final output.
However, in product discovery the degree of uncertainty is generally high. New information needed to lower uncertainty is sometimes too costly and time scarce. Hence, in product discovery heuristics rather than analytical decision-making can be strategic and efficient. This post provides with 6 heuristics, to begin with.
If you found this post useful, just share it with people that may benefit in your circle. If you want to know more about heuristics, or just continue the conversation, feel free to get in touch.