In this episode, we focus on moving from raw data to structured journeys by collaborating with AI to build journeys that are grounded in evidence and shaped by your expertise.

Why structure matters

A journey is more than a visual map. It’s a structured representation of the customer experience that:

• Helps you make faster decisions

• Reveals opportunities tied to real problems

• Connects insights to action

But turning raw, messy data into a meaningful structure is no small feat. That’s where AI comes in, giving you a massive head start by processing unstructured data and proposing an initial journey. The magic happens when your human perspective completes the picture.

Step 1: Start with the right input

Before you open your journey platform, you need input. This input typically falls into two categories: raw data and interpreted content.

Raw data is unstructured, qualitative, and comes directly from the source. Journey AI performs well with this type of material. Examples include:

• Textual input from surveys (open-ended, comment-style fields — not the quantitative questions)

• Interview transcripts

• Customer support logs

• Free-form feedback from tools or product reviews

These inputs provide rich detail, grounding the journey in real-world experience. For a deeper look at working with raw data, see our guide on grounding journeys in evidence.

Interpreted content, by contrast, is the result of human analysis and synthesis. It includes:

• Sticky notes from in-person workshops

• Miro boards organizing themes or insights

• Collaboratively drafted early journeys

• Research reports or presentation slides

This type of input provides early structure, fosters shared team understanding, and ensures alignment on language and themes.

Visual suggestion:

A side-by-side layout comparing "Raw data" (e.g., survey text, transcripts) with "Interpreted content" (e.g., Miro boards, workshop outputs), with arrows leading into a unified journey framework.

Both input types are useful: Raw data grounds your journey in real evidence, while interpreted content adds structure and team alignment. Knowing how to balance and use both sets you up for a smoother AI-assisted journey build. 

Step 2: Let AI do the heavy lifting

With your input ready, you can begin building your journey.

There are two approaches you can take:

1. Start from scratch:

AI reads your uploaded file (e.g., a CSV of open-ended responses or a TXT file of an interview transcript) and generates:

• A proposed journey structure, including phases and steps

• Insights that are automatically inserted into the relevant steps and tagged by type (e.g., pains, gains, needs)

2. Enrich an existing journey

If you already have a journey with predefined phases, steps, and insights (e.g., from a Miro workshop), you can copy that structure into TheyDo first.

To help AI perform better, enrich the existing journey by adding more context, such as descriptions for key insights or clearer step definitions. Then, AI can:

• Link relevant evidence from your uploaded data to existing insights

• Suggest new insights and automatically place them within the most relevant steps

Both methods provide a fast, credible starting point, but your input remains crucial in shaping a journey that’s both accurate and actionable.

Step 3: Collaborate with AI like it’s a new colleague

Think of AI as a junior analyst. It will do what it can based on the information it has. However, it lacks context: It doesn’t know your business, your audience, or what steps like "actioned" or "observed" actually mean in your specific journey.

That’s where your expertise comes in:

• Rename the journey meaningfully: Avoid generic names like "Member Experience." Tie the title to your business problem.

• Review the journey structure: Do the phases and steps reflect your intended focus?

• Check the mapping of insights: Drag and drop them to better-fitting steps. Expand insights across steps if needed.

This collaboration ensures the journey isn’t just visual, it’s powered by your experience and shaped by the specific context you understand best.

Step 4: Prioritize what to verify

With AI generation, you might end up with dozens (or hundreds) of insights to verify. Here, verification refers to the detailed review of AI-generated insights following your initial structure check in Step 3.

This involves evaluating each insight for accuracy, refining its language, and repositioning it within the journey as needed.

We'll explore this in greater depth in the next article, but first, it’s important to understand how to prioritize what to verify. Don’t try to verify everything at once; it’s rarely efficient or necessary.

As you continue ingesting more data, the top insights will naturally evolve. Each new upload reshapes the landscape, which means insights are not static snapshots. Prioritization helps you focus on what’s most relevant right now.

Depending on your needs, you can prioritize using one or more of the following approaches:

• Check pains first: These often point to urgent customer issues and business problems

• Focus on steps tied to key decisions: For example, "Membership Cancellation" or "Onboarding Activation"

• Start with top-scoring insights: Journey AI assigns an insight score based on reliability and impact (learn more about how scoring works)

These approaches help you make smarter use of your time and highlight what matters most. Among them, the insight score is often the most efficient way to prioritize, particularly when working with large datasets.

Step 5: Think ahead to scale and reuse

Once verified, your insights become more than sticky notes:

• They can be reused across other journeys

• They can be filtered, grouped, and queried

• They reveal and link to opportunities, helping teams prioritize actions and drive impact

Structure makes your insights portable, allowing them to be reused and scaled across various journeys. It also makes your journeys operational, turning them into live tools for decision-making and collaboration, where opportunities surfaced from insights can be systematically acted upon.

Key takeaways

• AI can accelerate your workflow, but it doesn’t replace your judgment.

• Your context, language, and priorities matter in shaping an accurate journey.

• Verifying and enriching your AI-generated journey is how you take ownership of the narrative, ensuring it reflects your team’s understanding and context.

• Structured journeys become living assets for strategy, decision-making, and collaboration.

What’s next

Want to go deeper? Watch the video walkthrough of Episode 3.

If you haven’t yet, check out:

Episode 1: Defining the Business Challenge

Episode 2: Grounding Your Journey in Evidence

Let your data lead the way, but don’t forget to steer.