How to run your own product discovery initiative

Discovery process overview

Up until this point in the blog series, the focus has been all about establishing the framework.

In this piece, I’ll dive into a qualitative discovery methodology that I’ve come back to time and time again for many different discovery initiatives, and that I truly enjoy experiencing. I’ll also give my guidance on how to set it up and stick to it, so you can go off and run the 8-week process step by step.

Here’s what I’ll cover:

A quick word on the process and the case study

This process is a mashup of principles and techniques, and there isn’t a specific name for it. The overarching framework for it was first introduced to me by my previous colleagues Jono Hey and Benjamin Merrick during my time at Nutmeg. Therefore, I cannot take credit, but naturally have adapted it to my own circumstances over time. Any inaccuracies or misinterpretations are due to me.

Discovery process overview
This is sourced from, a great site for really simple articulations of complex things.

One of the proudest moments of my career was launching and watching the evolution of digital wealth manager Nutmeg’s socially responsible investment portfolios, and it was a great example of how this type of discovery initiative can set you up for success. So I’m using it as a case study to help you see how it works in practice.

Identifying and answering the most important questions 

Launching a socially responsible investing product was something the Nutmeg team wanted to do for a long time. This is because it aligned perfectly with the company ethos and forward-thinking mindset.

It took a few years until the team felt the market for this product was mature enough. It wasn’t only a case of customer demand and alignment with the “product” side of the business, but the industry landscape itself was emerging at the time, and only when our investment team felt comfortable enough to construct and manage this type of financial portfolio on behalf of customers were we ready to start figuring things out concretely.

We were faced with a lot (!) of challenges for this product, and some of the questions this process helped us identity and answer were:

  • How do we build this product in a way that appeals to the widest range of customers?
  • What new data points and metrics do we introduce with these portfolios, and to the industry, that is useful, interesting, and easy to understand for customers?
  • Why are these specific data points more important vs others?
  • How do we ensure objectiveness, transparency, and alignment with our company’s core values and investment ethos alongside our other portfolio offerings?
  • How price-sensitive are customers for this type of product?
  • How do we communicate this product to different audiences, ranging from investment industry experts to average investors?

These questions represent the tip of the iceberg that the discovery process helped address. To be clear, there were loads of other challenges that internal teams outside of the discovery initiative had to solve for, from an investment, operational, and regulatory perspective that all needed to be aligned.

Launching a new type of product, in an emerging industry, that by nature is extremely subjective and opinionated is very challenging. Going through this process gave us confidence in those early days about what we were building and how we were going about doing it.

The 4 stages

There are 4 primary stages involved. These are not always sequential. As in life, things go back and forth, and are rarely in perfect order. And that’s OK!

Stage 1. Observations

This stage is where the fun starts. All of the admin and set-up related work to get to this point is crucial to your success, but it just doesn’t feel as good. If you haven’t read any of the previous articles, start at the beginning to make sure you’ve put in the necessary groundwork.

By this stage, you should be:

  • Finalizing the types of people you need to interview and recruiting them
  • Beginning to conduct the interviews
  • Getting a better understanding of the topic
  • Sifting through customer interview recordings and notes to increase familiarity
  • Conducting prototype tests
  • Creating customer profiles

This stage sits in the “concrete” phase because the observations you make are based on identifiable observations throughout the interviews, by what people said and did within the interviews.

Conducting interviews should follow a structured approach. I always address this by creating a customer interview guideline, with a tailored question bank. This helps to make sure everyone’s asking the right type of questions in the right way, for myself and others that are conducting the interviews. This ensures consistency across interviews.

In these guidelines, I always put reminders for myself as well. For example, seemingly simple things like not forgetting to keep your phone charged and to enable audio recording, asking for formal permission to record them, getting the research reward voucher receipt signed, and so on. These reminders have saved me countless times when I have been on the move. 

Even the way that interview notes are taken should be standardized. This way, aggregating and analyzing these notes later on by different members of the team doesn’t turn into a messy project. Also – and trust me on this – it’s much easier to understand notes taken by a different person a few weeks previously if they follow a standardized format. Things like taking direct quotes, using shorthand expressions, and other tips help a lot with the note taking process.

When creating customer profiles – guess what? Those should be standardized formats as well for easier creation and analysis during exercises later on. 

If possible, it’s great to be able to take over a room so that your team can put up the customer profiles on large pieces of roll-out paper on the walls for a period of time. This gives people in your organization the opportunity to drop by, get more familiar with the interviewees, and see what’s going on in the process. This can help with stakeholder management later on. 

Stage 2. Insights 

This stage is about transforming those observations into insights. You’re entering a stage where you’re beginning to conduct different types of analysis and apply abstract thinking techniques.

Being familiar with the customer profiles created in Stage 1 (some might spill over into this stage, remember that’s OK!) is crucial in order to come up with insights. Unless you have taken the time to digest all of the interviewee’s information, it will be difficult to realize conjecture points such as emerging trends, outlier problems, needs vs wants, and areas you’re uncertain about and need to focus more on.

You can also overlay the findings from your interviews with secondary research during this stage, whether from past discovery initiatives or external research sources. Additional information is never a bad thing at this point. 

There are a variety of exercises you can conduct to get the brain juices flowing to help assess information from different perspectives. Some common exercises I apply are:

  • Thematic and emerging themes grouping
  • 2×2 axes analysis
  • User journey and mind mapping
  • Needs mapping
  • Raising new questions 
  • Powers of ten

There are countless exercises a team can use to analyze customer interviews as a group or individually. Use the ones you’re familiar with and experiment with new ones. Apply a few different techniques to get different perspectives of the information and the problem.

Remember – this is a team exercise, and secondary stakeholders should participate here. It’s great to be able to carve out a few hours at a time to conduct these exercises, ideally in the same location as the interview profiles are for reference. As a rule of thumb, aim for 2 hours at a time. You’d be surprised how fast time goes by and how long it takes to digest information. An example of the themes grouping exercise I mentioned above is outlined below:

  • Having the profiles introduced and explained by the people who conducted and participated in the interview 
  • Giving everybody time, e.g. 30 minutes, to walk around the room and familiarize themselves with the profiles 
  • The leader of this exercise should have some broad categories defined and posted into different sections on a wall for people to potentially align on and add to
  • Having everybody begin to identify common themes, and create new ones, between the interviewees, writing them on post-its, and grouping them on the wall together
  • Having the group assess each other’s insights – it is often at this point where emerging trends will start to get surfaced
  • Conducting group analysis exercises to collect related themes

Remember – it is the primary stakeholders’ job to explain and guide the group through this process in addition to participating.

This is also a good stage to start sharing with the wider company. People might be interested in what is going on in the room with all of these profiles, so let people know what’s going on! Share with the company where you’re at in the process, some interesting findings you’ve surfaced so far, the challenges you’re facing, and the next steps. This helps keep people engaged.

Stage 3. Principles

This stage is about creating the design principles your team needs during the creation of that feature or product. It’s the set of guidelines to make sure your team is on the right track.

For example, if during the interviews a trend was realized that all of the participants are primarily researching about the security aspects of the product before making a purchasing decision, you should be incorporating that insight into your product in the relevant places, such as public-facing content, onboarding, reporting, and so on.

It’s common for companies to only focus on building trust at the very forefront of the product, such as on product landing pages and marketing content. In reality, building trust in products that people have a reliance on for a crucial part of their lives, such as their investments, happens gradually. This generally happens at deeper levels of the product or service, where a large bulk of building trust actually happens. This is especially true for services that are based on retention and consistent actions. 

Drawing on my example of creating the SRI portfolios at Nutmeg, a couple of major themes the team was noticing from the insights generated in the interviews were:

  • People wanted their portfolios to have a better impact on society and/or the environment, but didn’t exactly know how that could happen through investing themselves or what that impact should be
  • There were many different views, levels of understanding of the industry, and degrees of sensitivity in this realm across investors
  • The wide range of investors just wanted to have a feeling of doing “better” in general through their investments vs having a specific area of focus
  • People said they were willing to trade off on price (e.g. investment returns) a bit, but when going through testing of different thresholds, what they said and what they did varied dramatically
  • Showing the impact that a greater collective of people was having, rather than just individual impact, was significantly more meaningful and rewarding and would lead to further engagement. This got us thinking about how and why movements get created and the impact they can have.

Here are some of the principles we created to guide us when creating this product:

  • Show the collective impact of investments
    Certain reporting features were tailored to communicate the collective impact of investors. Blogs were created and other metrics tracked to show this consistently as time progressed.

  • Demystify the tradeoffs
    What was great for us was that, in the end, the base pricing wasn’t more expensive than other portfolios. In reality, is it really crazy to believe that companies that conduct ethical governance practices, create sustainable processes and hold their company to higher standards have good results? We communicated this by comparing return profiles of different investment profiles in fair and objective ways.

  • Getting concrete about the ends and downplaying the means
    Create easy to understand metrics that show clear impact. We assessed over 100+ data points and whittled them down to 10, representing meaningful information for customer portfolios to keep it simple at launch. Most people just wanted the results and weren’t interested in the nitty-gritty details of how we selected and managed assets. We made that type of information accessible to those who wanted it, but didn’t put it front and center.

Other example principles I have seen in the past:

  • For target audience A, ensure you communicate about the technology and security of the service
  • For target audience B, talk about the quality and expertise of people within the team
  • Continuously build trust throughout process X by making the customer go through actions A, B and C overtime before trying to up-sell 

This type of information is critical for teams when creating or updating a particular digital experience for a customer. It gives the team tick boxes to check off for each step of creation, helping to ensure things are going in the right direction. It produces consistency across different areas of the product. It acts as the glue between all the teams working on the same thing, in different capacities, keeping everything together.

Stage 4. Solutions

At this point, your team has probably started thinking of potential solutions. This is great, but don’t rush to think that you’re done. Remember – these ideas likely haven’t been validated enough and need to go through a few iterations and feedback loops.

This is where your product development process comes into play. What you have done though is put forth potential versions based on validated insights. This is another great accomplishment because it represents a significant milestone and is worth communicating across the wider company.

When presenting, I always try to give context about the process and the steps involved, the different interviewees, the insights and principles established and example prototypes being tested. You might be surprised about how interested people across the organization are in the work that’s being done here. 

Earlier in the article, I mentioned that these stages are not always sequential. Chances are that you’ve jumped back and forth between creating insights and principles. The same will apply here. You will find yourself jumping back between the solutions and the previous steps. This is natural.

Things to do before launching your own discovery initiative 

As with everything, you want to maximize your chances of success by doing some preparation.

First, you need to know what it is you want to conduct discovery on. Chances are, you are reading this article with something in mind already.

The good thing about this discovery methodology is that it’s great in multiple scenarios. Here are just some examples I’ve used it for:

  • Launching a new product based on customer feedback and evolving markets
  • Launching a new feature based on direction coming from top-level management
  • Getting a better understanding of specific target audiences
  • Addressing how to market and continually communicate a product for specific audiences

Chances are, your company has its own prioritization process, so this is a great method for validating those priorities and turning them from an idea into something concrete.

Who you will need to be involved

To make this work, you’ll need two types of participants: primaries and secondaries.

Ideally, you have at least two primary participants to drive the process over the weeks. Secondaries are involved more loosely. It’s critical you have a core group of people involved throughout the entire process to be innately familiar with all of the different parts and to act as the glue.

This is important because when it comes time for analysis and ideation stages, unless you are extremely familiar with all of the different inputs you risk not being able to efficiently piece different insights together. Below is a table representing the different responsibilities involved in the process.


Timelines, timelines, timelines.

As with any project, stakeholders will always want to know timelines. This is a fact of life. Below is a rough guideline (I always err on the conservative side) with steps and milestones.

If this is the first time your team is conducting this type of initiative, expect specific parts of the process, such as customer interviews, to take longer if the infrastructure is not in place.

The process also assumes that there is alignment on the objective of the product discovery initiative, the expected outputs, and some initial recruitment has been started for primary and secondary stakeholders.

In general, before this type of initiative kicks off, I like to give an overview in separate sessions about what this process looks like, the methodology, and example outcomes. This is where you can get your eye on people that might be interested and start recruitment from different teams.

I’ve broken the image above into weeks and stages, with detailed guidance on roughly what should be happening that week. Try your best to carve out time to make sure these things can tick along. 

Week 0 prep

Week 1-2

In my experience, there is always some spillover from what should be done prior to officially starting. It happens and it’s OK.

In the first week, as long as you have at least one other primary stakeholder, the objectives and expected output of the initiative, and can begin reaching out to interviewees, you’re in a good place already. As a bonus, if you can get your customer interview guidelines finalized and any tests you want to conduct during your experiment sorted, even better.

Remember that, during weeks 1 and 2, you’re in Stage 1: Observations. This is where you’re conducting interviews in addition to any spillover work such as getting lists of potential interviewees together and refining interview guides.

Week 3-4 

Chances are that your team might not have finished all of the interviews yet. Regardless, this is a good time to start building your interviewee profiles.

Build each profile with the other persons involved in the interview – you’re firmly in Stage 1-2 territory here. Ideally, each profile should be built immediately after the interview, when information is fresh in your mind and your notes are on hand. In reality, this isn’t always feasible, and it’s fine if it happens a little later.

Start getting these profiles up on the wall, or wherever they can be accessed.

Enjoy this stage – things are really starting to get interesting!

Week 5-6

In Week 5, hopefully, you have completed the majority of interviews. Even if you haven’t, if you’ve done at least 80% of them, I’d recommend getting firmly onto Stage 2: Insights. Momentum is important in this type of project. I can’t quantitatively justify why momentum is important, but it is!

So Stage 2 is about creating those juicy insights. This is when you’ll be conducting those analysis exercises to identify trends, outliers, problems and opportunities, and any other insights that are being surfaced from the interviews. 

I’ve been in situations where 80% of the interviewees are saying the same thing about a particular topic. However, there are one or two strong outliers from the other 20%. In this case, you should recruit a few more interviewees to “smooth out the data” and confirm whether that 20% really is just an outlier, or if it’s actually a trend. 

During these weeks, depending on if you feel you have some clear insights that you can act on, you might start to enter Stage 3: Principles

Week 7-8 

Stages 3 and 4 happen within this time frame. Stage 3 is all about creating those principles to guide you through your process of creating solutions. The output of these principles should resemble checks for your team during the solution, such as:

  • Be transparent about the pricing
  • Highlight the security features
  • Communicate the social impact of this investment
  • Don’t push one offering over the other to the potential customer

This is also the perfect stage to share your learnings. At this point, you’re able to talk about:

  • The overall process and experience of the initiative
  • Share details about the interviewees
  • Communicate the learnings
  • Talk about next steps and potential solutions

Stage 4: Solutions begin here as well. In reality, you probably won’t have something concrete to put into your roadmap for development just yet, but you will have the guidelines to make it, and potentially a proof of concept. It’s all about iterating consistently, after all.

Regardless, while you’re sharing your findings, having these example proof of concepts is great to share internally. It helps to put meat on the bones for the topic of the initiative because it’s something tangible – having visualizations is more powerful than just having a bullet point list of insights and principles. Get them into something that could be reality!

Closing curtains…

The end-to-end timeframe for this type of initiative can range from 6-8 weeks. While it’s hard to give concrete timeframes for this type of work, it’s important to keep track of product development purposes.

This is on the basis of having a basic product discovery infrastructure established and one or two persons familiar with the process. This type of initiative takes a considerable effort, dedication, and should not be underestimated.

Having said that, any organization with the fundamentals in place can do this in addition to implementing the framework described in previous articles. You don’t need a fancy user research team with complicated processes to do any of this!

Please don’t hesitate to reach out to me with any questions. I would love to hear your feedback. I aim to continually update this blog series based on the questions I receive from readers to make it more valuable over time.

Good luck!