Go to app

Prix fixe to a la carte: tailoring insights with Okta’s Jared Forney

26 May 2024
Jared Forney

Get the low-down from Jared Forney about sharing insights across your organization. Jared is the Research Operations Principal at Okta and spoke at Insight Out—Dovetail's first global conference for product teams.

I’ve been working on a framework that tailors insight delivery. For the last two years, I’ve been focused on Okta’s research operations. I have a number of responsibilities including knowledge management, governance, recruitment coordinations, and research workflows.

In the seven years I’ve been there, I’ve watched the organization grow and mature from a research standpoint to a product organization of 900 in a broader company of 6,500 globally. In that time, we’ve been using research repositories as a function for about three years. And we’ve been happy Dovetail customers for about two years.

We’ve noticed some changes in the evolution of our research maturity and how it’s changed our relationship with insights and how we deliver them.

Try the new magic features

Ask questions of your data, get AI-suggested highlights from customer calls and more with the new magic features.

Get Dovetail free

1. Understanding the problem

Researchers find Dovetail a powerful tool for analysis and development. But feedback from some of our stakeholders told us two key things:

  • Our stakeholders were having trouble finding insights in our repository. We were also starting to notice increasing evidence of artifacts with insights being created outside the repository in external tools, for example, in Google Workspace.

  • Our researchers were spending more time digging for insights in our repo. And this sometimes resulted in duplicate work when serving that first use case and generating those other artifacts for presentations or readouts. This didn’t leave a lot of time amid their compressed study schedules to connect the dots between individual studies.

Faced with these two fundamental challenges regarding the current state of our repository, I started to think about our research organization like a restaurant. I realized our researchers are our chefs. They’re taking all these dishes in the form of our insights and serving them to our stakeholders. And there’s this ongoing relationship and cycle formed with all our researchers when they’re serving insights throughout our organization.

Then I started to think about this as a kind of loose framework. I held a series of listening sessions and spoke to folks throughout our product organization, our researchers, of course, as well as designers and product managers. I wanted to get a fundamental understanding of the problem.

I came away with three big things:

  • Our chefs—our researchers—were faced with an overwhelming number of orders.

    You can never have enough researchers or research in an organization.

    There’s always an ongoing demand.

  • There were too many different dishes our chefs had to choose from and prepare. There were a lot of different ways to generate insights within our organization and our repository tool. And there was a challenge with consistency in that delivery. This made it hard both on the researcher front to generate those insights and also on our diner—stakeholder—front in finding those insights after they were made.

  • We had “VIP diners,” the folks who come in last minute without a reservation, asking for off-menu items, things we weren’t necessarily equipped to deliver under our model with respect to research insight.

We also realized a repository alone wasn’t the sole solution.

When we choose a repository tool, we often have stars in our eyes and envision it’s going to be the centralized place where everybody is going to go for insights all the time. But what we realized at Okta was that our research repository was like a pantry full of ingredients and some recipes to follow, but our diners didn’t necessarily want to cook.

We realized we needed to rethink our relationship with insights, how we deliver them to our stakeholders and that collaborative model.

This new framework is composed of four key pieces:

  1. What makes a great dish? What makes for a great insight, first and foremost?

  2. Who are our diners? We know they’re the people we work with every day, our stakeholders, but we need to understand their needs, how they order, how they need those insights delivered, and how this differs from the approach we’re taking today.

  3. How do they order? How do they need research insights delivered?

  4. When do they need it? This is relating to time and context, which may differ based on the project, the timeline, and even the stakeholder.

With that framework and those four things in mind, let’s look at what makes a great dish.

2. What makes a great dish?

What is an insight? This varies from organization to organization.

At Okta, we wanted to focus on a common format and a uniform deliverable to get the most out of delivering learnings throughout the organization.

Having a common format for insights also drives consistency and quality, much like having a consistent standard for a dish helps chefs to cook more effectively and efficiently. The dish is consistent; it comes out the same way every time.

From that standard and consistency, we also set expectations with our stakeholders, our diners. So when dishes are consistent in a common format, they’re efficiently delivered and ultimately easier to digest.

At Okta, we’re thinking about this construction of insights like a mise en place. What do you do before you cook? You arrange all the items you’re going to need to make that dish effectively to minimize waste of movement and resources.

For us, it comes down to three key things for an insight:

  • The context: What the insight is about, what features, what part of the product, the audience, etc.

  • The recommendation: Our stakeholder feedback showed they view our researchers as subject-matter experts and look to our researchers to provide guidance for what to build and where to go.

  • The voice of the user: This is one of Okta’s core values—to love our customers. Another thing we heard loud and clear from our feedback sessions was that the voice of the user was super-important. Having a customer quote to solidify and ground that insight was also fundamental to an insight for our organization.

So, we’ve got our ingredients, we’ve got our dishes. Now, let’s start thinking about planning a menu.

3. Plan your menu

When you’re planning these menus, there are three main considerations:

  • The needs of your stakeholders (diners) and your capabilities to serve those needs. There’s a whole range of different deliverables, insights and depth you can take as a researcher or research organization. Sometimes it’s best to start small to ensure you can deliver things effectively.

  • Meet your diners where they are and the context they’re in. We discovered a lot of our stakeholders are spending time outside Dovetail and don’t necessarily view it as a first-order destination. So they’re living in tools like FigJam, Slack, Google Workspace, Tools, and Confluence. It became important for us to be able to match that and meet people where they are.

  • Stay flexible. As we roll out a new framework like this, we anticipate making adjustments and changes over time. So, flexibility is a really important value, just as in any aspect of collaboration.

Prix fixe menu

As the name suggests, this is a fixed-price menu, offering a set order of items for a single price. It’s a menu with a clear beginning, middle, and end.

There are a lot of benefits to this, particularly if you carry out a more generative, exploratory style of research. It’s easy to control for cost and it’s replicable. You’re doing the same process each and every time.

The other advantage of a prix fixe menu is that it’s chef-curated, or, in this case, researcher-curated. It allows the researcher to control the experience and shape the way insights are delivered in a more set way.

Given that fixed nature, there are also drawbacks. It’s not very adaptable, as its name suggests, and it takes more time generally to prepare these larger, fixed setups. It may not be suited for something more tactical.

An example of a prix fixe menu for a large, foundational generative study is:

Literature Review Featuring internal and external sources

Research Question Design With stakeholder workshop

Customer Interviews Including clipped session debriefs in Slack or another tool

Research Readout Conducted live with feature team, including Q&A

Post-Research Summary Accessible in repository with responses from Q&A

This approach would be largely researcher-driven regarding how the insights are formulated and how it’s structured. And, again, it has that clear beginning, middle, and end in a repeatable process.

A la carte menu

This approach is more about selecting items individually rather than them being in a set order.

The obvious benefit of this approach is flexibility and choice. The menu is not always defined in the same way every time and the ability to mix and match is its strength. But you have to be careful not to combine too many things that may not necessarily fit because, ultimately, the cost can wind up greater than if you chose another method.

In practice, this allows you to break down different forms of insight delivery for a more targeted option.

An example of an a la carte menu in the content of insight delivery is:

Voice of the Customer Insights summary – 2–3 days Customer quote/clip – 1–2 hours Customer AMA prompt – 1–2 weeks

Product Improvements Highlight reel – 1–2 days Usability review – 3–5 days Using custom admin – +1–2 days

New Product Development Desk research report – 4–5 days New concept evaluation – 2 weeks

These are things that could be done more on demand, differing from a fixed-price option in a set series of steps.

Hybrid menu

This menu framework makes hybrids possible, so we don’t have to stick to the set menu options.

The food-ordering approach of omakase, meaning “I leave it up to you,” is a great example. This is a variation on the prix fixe method where there might be a set sequence of dishes that change based on seasonality. This is important for accommodating the diner’s taste or what’s fresh from the market.

In the case of Okta, the biggest time of year is conference season. Our Oktane conference every year is a great example of when we might use this omakase method. We adjust our framework for how we deliver insights, for example, with feeds, tags, or attributes in Dovetail that focus specifically on the Oktane event because it’s top of everyone’s mind as our big customer event of the year.

Set menu

This is the food-truck model for more resource-constrained setups, such as research teams. This can be a great option to focus your insight delivery into one thing.

What’s one thing you do really well? For a food truck, it might be a great chicken sandwich. This approach is focused on efficiency and the highest quality for a narrowly defined insight.

4. Applying menus to research strategy

We’ve talked about the importance of fresh ingredients and dishes and setting up menus in an effective way. How do you apply this to research strategy as a whole?

Aren’t these menus just playbooks? In a way, they are, but there are two key differences between them and a traditional research playbook:

  • They’re modular. Playbooks tend to be large, single-point, monolithic structures. They also tend to be more referential, which is great if you’re handing something off to someone else or training someone, but they’re tricky to update and apply in a point-based, piecemeal way. This is where menus come in because they’re easy to spin up and adopt.

  • They’re service-centric. Playbooks tend to be more for the person writing it—the operations or research professional. Menus can be more directed at the stakeholders, and they involve them more in the delivery process.

In summary

Menus are flexible and adaptable. They can be used throughout research organizations to drive insights more effectively, efficiently, and repeatably.

What’s fundamental to menus as a whole is that consistency is key to a great dish. Having a fundamental common agreement on what an insight is in your organization is crucial to getting menus right.

Different restaurants (organizations) need different menus. I’ve worked in small research organizations and bigger ones. We don’t have to shoot for a four-star Michelin experience every time.

For example, the US burger chain In-N-Out can be a really great experience. They have a great menu, and nobody is going to compare a top-class restaurant like The French Laundry and In-N-Out. But they can both be great experiences for the customer, depending on their goals and how the restaurant delivers on that experience.

No matter the size of your research organization or its stage of development, menus can sharpen your insight delivery and make sure you’re giving your stakeholders the best possible experience.

Editor’s note: This article is a condensed overview of Jared Forney’s remarks and live Q&A session at Insight Out 2024.

Keep reading

See all

Decide what to build next

Decide what to build next

Get Dovetail free


OverviewChannelsMagicIntegrationsEnterpriseInsightsAnalysisPricingLog in


About us
© Dovetail Research Pty. Ltd.
TermsPrivacy Policy

Log in or sign up

Get started for free


By clicking “Continue with Google / Email” you agree to our User Terms of Service and Privacy Policy