Continuous Discovery
Continuous discovery refers to the iterative process of identifying and understanding customer needs, problems, and opportunities on an ongoing basis within the context of software product development. By continuously gathering and integrating customer feedback into the product development cycle, teams can ensure that the product evolves in alignment with user demands and market trends.
Purpose
Up to 90% of the features we build fail to deliver the expected value. The primary purpose of continuous discovery is to reduce the desirability risk of building the wrong product or features. There are no right or wrong answers when trying to decide what features to develop, only better or worse ones. Continuous Discovery ensures that decision-making is informed by up-to-date insights from real users.
Principles and Practices
Effectiveness
| Principle | Goal | Practice |
|---|---|---|
| We don't know our customer's real problems | We need to ensure we are solving real customer problems and not whiteboard problems. | By conducting regular generative research, using both primary and secondary sources capturing both qualitative and quantitative data, we can generate insights on customer behaviours. |
| Customers know they have problems. They don't necessarily know the solutions | We need to identify all of the customer pain points and unmet needs. | By aligning the research methods used with the ladder of evidence we can move past opinions to get real insights. |
| Not all opportunities are universal | We need to ensure that we are building solutions that are valuable to cohorts of customers and not just solving issues for a single customer. | By documenting the customer journey and mapping opportunities onto it we can identify recurring patterns across customers. |
| Not all opportunities are worth solving | We need to identify the opportunities that align with our strategy and deliver the best impact for customers. | By prioritising opportunities we can ensure that we are focussing on the opportunities that may have the biggest impact on customers and the business. |
| Unused work provides zero effectiveness | We need to ensure that research work is understood and influences product development. | By embedding designers and developers into the research practices we can ensure that insights are distributed around the team, that the nuance of the research is understood and the diverse involvement also helps to identify many, and more diverse, product opportunities. |
Efficiency
| Principle | Goal | Practice |
|---|---|---|
| Work expands to fill the time available for its completion | We need to balance the cost of research with the insights that can be gathered because research has diminishing returns. | By setting strict time limits on research objectives we can align the investment with our current priorities. |
| Not all work is equal | We need to maximise the amount of time that stream teams are gathering insights rather than completing admin tasks. | By automating research activities as much as possible we can lower the cost of research. |
Sustainability
| Principle | Goal | Practice |
|---|---|---|
| People leave teams | We need to minimise the amount of knowledge lost when people leave the team. | By creating an accessible research repository we ensure that anyone can access the research results at any time. |
Criticisms
Despite its benefits, continuous discovery faces several valid criticisms.
| Criticism | Description | Mitigation Strategies |
|---|---|---|
| We already know what the customer wants | Companies have large backlogs of work. We would be better diverting resources to delivery rather than discovery. | Highlight how often features fail to deliver the expected value. Building less, but more effective features, will deliver better results. |
| Resource Intensiveness | Continuous discovery requires significant investment in time and resources to engage with users and analyse feedback. | Use research objectives to prioritise discovery activities, and use lean research methods to optimise resources. |
| Analysis Paralysis | The volume of data and feedback can overwhelm teams, leading to indecision. | Enforce continuous decisions by shifting from a large upfront research cycle to continuous discovery. |
| Bias and Misinterpretation | There's a risk of introducing personal biases in selecting users for feedback or in interpreting the data. | Use diverse and representative user groups for feedback and involve the full Stream Team in data analysis. |
Anti-patterns
To avoid pitfalls in continuous discovery, be wary of these anti-patterns:
- Skipping Discovery: We think we know what problems customers are facing so we skip discovery to save time. But this is why 90% of features fail to deliver the expected value.
- Echo Chambers: Relying on a narrow, homogeneous group of users for feedback may result in a product that doesn't cater to the broader market.
- Overvaluing Quantitative Data: prioritising quantitative data (like analytics) over qualitative insights (such as user interviews) can miss the nuanced understanding of user needs.
- Ignoring Negative Feedback: Focusing only on positive feedback and dismissing criticisms can hinder the identification of crucial areas for improvement.