Teresa Torres on how to interview customers, automating continuous discovery, the opportunity solution tree framework, making the case for user research, common interviewing mistakes, and much more

Jun 30, 2022 Episode Page ↗
Overview

Teresa Torres, author of Continuous Discovery Habits, discusses the Opportunity Solution Tree framework and continuous discovery. She shares how to integrate customer conversations weekly, improve interviewing skills, and shift from output to outcome-focused product development.

At a Glance
14 Insights
47m 32s Duration
13 Topics
5 Concepts

Deep Dive Analysis

Introduction to Teresa Torres and Her Impact

Understanding the Opportunity Solution Tree Framework

Netflix Example of an Opportunity Solution Tree

Challenges in Identifying Customer Opportunities

Implementing Discovery in a Feature Factory Environment

Defining Continuous Discovery and its Importance

Addressing 'No Time for Discovery' Objections

Automating Customer Recruitment for Weekly Interviews

Maintaining Unbiased Solution Exploration and Team Collaboration

Best Practices for Customer Interviewing

Common Mistakes in Customer Interviews

Adapting Discovery Practices as Companies Grow

Differentiating User Research and Assumption Testing

Opportunity Solution Tree Framework

A simple visual decision tree that starts with an outcome, branches into opportunities (unmet needs, pain points, desires), then solutions, and potentially assumption tests. Its purpose is to provide structure for product teams to navigate the complex problem of starting from an outcome and determining what to build, moving away from a feature-focused approach.

Opportunity (in OST context)

An unmet need, pain point, or desire that is derived from customer stories. It should be framed specifically enough to be solvable, distinguishing it from a solution, and its identification requires careful interviewing to uncover needs customers may not even be aware of.

Continuous Discovery

The practice of continuously incorporating customer input into the process of deciding what to build. It acknowledges that digital products are never truly 'done' and require ongoing iteration, improvement, and continuous feedback loops to make better product decisions over time.

Product Trio

A well-functioning team comprising a product manager, a designer, and a software engineer who collaborate closely. This trio works from a shared understanding, reducing disagreements and continuously seeking the best options, rather than operating in functional silos with internal politics.

Assumption Testing

An activity that helps evaluate a solution by breaking it down into its underlying assumptions, prioritizing them, and running small, rapid tests for each. This is considered the start of delivery and is crucial for making continuous discovery sustainable by quickly validating or invalidating parts of a solution.

?
What is the Opportunity Solution Tree Framework?

It's a simple visual decision tree that helps product teams manage the complex problem of starting from an outcome and figuring out what to build, by structuring the process from an outcome to opportunities, solutions, and assumption tests.

?
How should product teams approach finding opportunities for the Opportunity Solution Tree?

Opportunities emerge from customer stories, which are best collected by asking customers to recount specific past experiences rather than asking direct, out-of-context questions about preferences or hypothetical actions.

?
What should an individual contributor do if their company operates like a 'feature factory' and doesn't prioritize outcomes?

Focus on changing individual work habits by finding customers to talk to and understanding how current work contributes to business outcomes, even if solutions are prescribed, rather than trying to force organizational change.

?
What is continuous discovery?

It is the practice of continuously incorporating customer input into the decision-making process of what to build, recognizing that digital products are constantly evolving and require ongoing feedback loops.

?
How can product teams address leaders who claim there's 'no time for discovery'?

Frame discovery as continuous, not project-based research, and integrate it in parallel with delivery. Start with small, consistent discovery efforts (e.g., one interview a week) to gradually improve product bets rather than halting delivery for extensive research.

?
How can product teams automate the process of recruiting customers for weekly interviews?

Allow customers to opt-in while using the product (e.g., via an in-product prompt asking if they have 20 minutes to talk) or leverage internal teams like sales or support who are already in contact with customers.

?
How can a PM stay unbiased about a potential solution and foster better team decision-making?

Work with multiple solutions for the same opportunity to compare and contrast, and foster a collaborative environment within a 'product trio' (PM, designer, engineer) where shared understanding minimizes disagreement and encourages seeking better options.

?
What is a key best practice for interviewing customers?

Focus on eliciting detailed stories about past experiences by asking 'what happened next?' and being genuinely curious about the timeline and specific actions, rather than peppering them with a long list of direct questions.

?
What is a common mistake people make while interviewing customers?

Staying shallow in conversations, not digging into moments of friction or unmet needs, and focusing on what people *would* do or *think* rather than what they *actually did*.

?
How does discovery change as a company grows?

The fundamental unit of an empowered product trio working towards an outcome remains the same, but larger companies require more lateral collaboration with adjacent teams to manage dependencies and ensure a coherent user experience.

?
When should you run an experiment versus rely on user research?

User research (qualitative interviewing) is for learning about the opportunity space (unmet needs, pain points). Experiments are a form of assumption testing, where specific assumptions underlying a solution are tested, often in small, rapid ways, as the start of delivery.

1. Adopt Continuous Discovery Habits

Integrate customer conversations weekly by automating recruitment, making it easier to do than not, to continuously make better product decisions. This shifts from project-based research to an ongoing process.

2. Utilize Opportunity Solution Tree

Structure product strategy visually from an outcome to opportunities and solutions, helping teams manage the complex problem of deciding what to build. This framework adds structure to a messy problem, enabling more strategic decisions.

3. Master Customer Interviewing

Elicit rich customer stories by asking ‘what happened next’ and being curious about their experience, rather than direct questions, to uncover unmet needs and pain points. This approach leads to more reliable insights and uncovers needs customers may not even be aware of.

4. Empower Individual Discovery

If your company is a ‘feature factory’ or restricts customer access, find ways to talk to customers yourself, even outside official channels, to gain context. This allows you to make better daily decisions and develop a deeper understanding of customer needs.

5. Shift from Project to Continuous

Integrate discovery and delivery in parallel, rather than as sequential phases, to continuously make better bets over time. This approach ensures you’re always learning and improving, even if leaders claim there’s ’no time for discovery’.

6. Automate Customer Recruitment

Allow customers to opt-in for interviews while using your product (e.g., via an in-app prompt) or leverage internal teams (sales, support) to schedule conversations. This makes weekly customer conversations effortless for the product team, who only needs to show up.

7. Deconstruct Opportunities

Break down large, evergreen opportunities into smaller, addressable needs as you move down the Opportunity Solution Tree. This allows teams to tackle manageable problems while still contributing to bigger, harder challenges.

8. Work with Multiple Solutions

When tackling core product functionality or differentiators, explore and compare multiple solutions for the same opportunity. This guards against over-committing to a single, potentially suboptimal, idea and helps make better decisions.

9. Embrace the Product Trio

Foster deep collaboration between the product manager, designer, and software engineer to make joint decisions from a shared understanding. This reduces disagreements and leads to building better products by leveraging diverse perspectives.

10. Prioritize Assumption Testing

Break down ideas into their underlying assumptions, prioritize them, and run small, rapid tests (half a dozen to a dozen per week) to evaluate solutions efficiently. This makes continuous discovery sustainable and allows for comparing multiple solutions simultaneously.

11. Measure Solution Impact

Instrument your product to measure the impact of everything you release, rather than assuming no risk in bets. This helps you hone your judgment on where risk lies in new ideas and solutions over time.

12. Address Small Data Skepticism

Counter concerns about making decisions based on small data by highlighting that one interview is better than zero, and that small experiments are valid due to continuous feedback loops. Product teams are in the business of changing behavior, not just seeking new knowledge, and can get large-scale data through live production prototyping.

13. Over-Index on Discovery Initially

When new to discovery, intentionally do a little too much research to develop your judgment for assessing risk in product bets. This helps build confidence and skill in identifying where robust discovery is most needed.

14. Unlearn Business Silos

Challenge the notion of functional territories and power dynamics in decision-making, instead focusing on collaborative ‘doing’ to build better products. This fosters a more effective working environment where teams intuitively collaborate rather than engaging in internal politics.

I think interviewing is a grossly underestimated skill. Grossly underestimated skill.

Teresa Torres

We're so used to like everything being mediocre. We're not even aware of a lot of these needs that we have. But when we tell our stories, especially if you start to train your ear for this, you start to hear those needs.

Teresa Torres

If your interview feels like you're having a beer with a buddy, that's a good sign. Like it should be that casual and that conversational.

Teresa Torres

Every human in business is making decisions with zero data. So I'm going to go with one is better than zero.

Teresa Torres

The real measure is tell me about your behavior. What did you actually do?

Teresa Torres

Automating Customer Recruitment for Weekly Interviews

Teresa Torres
  1. Allow customers to opt-in while they are actively using your product or service, for example, through an in-product prompt asking if they have 20 minutes to talk.
  2. Upon a customer opting in, automatically send them scheduling software (e.g., Calendly) so they can pick a convenient time on your calendar.
  3. For B2B buyers and decision-makers, leverage internal teams (sales, account managers, support) who are already in regular contact with these customers.
  4. Define a specific trigger or target profile each week for internal teams (e.g., 'this week, we're looking to talk to somebody experiencing this particular need or pain point').
  5. Instruct internal teams to use scheduling software to book interviews directly onto the product team's calendar, ensuring the product team is not involved in the recruitment process itself.

Effective Customer Interviewing

Teresa Torres
  1. Begin by asking the customer to recount a specific, recent story related to their experience with the product or problem space (e.g., 'Tell me about the last time you watched something on a streaming entertainment service').
  2. Maintain genuine curiosity and use open-ended follow-up questions like 'What happened next?' to encourage the customer to elaborate on the timeline, context, and specific actions of their story.
  3. Actively listen and periodically summarize what you've heard to demonstrate understanding and to guide the conversation back to moments or details that require further exploration.
  4. Focus on collecting information about the customer's actual past behaviors and experiences, rather than asking about hypothetical actions, opinions, or why they think they do something.
  5. Strive for a casual, conversational tone that feels natural, akin to 'having a beer with a buddy,' rather than adhering to a rigid, question-by-question interview protocol.
11,000
Number of students taught through Product Talk Academy Approximate number of students Teresa Torres has taught, just about to cross this mark.
Hundreds
Number of product managers coached by Teresa Torres Excluding those taught through courses, focusing on direct coaching.
3-7
Recommended number of top-level opportunities in an Opportunity Solution Tree For cognitive processing, similar to Miller's magic number.
One interview per week
Minimum frequency for customer interviews in continuous discovery To make continuous discovery sustainable.
3
Recommended number of ideas to work with simultaneously for assumption testing To allow for comparison and contrast of solutions.
Half a dozen to a dozen
Number of assumption tests a proficient team can run in one week When methods are shifted to be small and rapid.