Expert Interview

The following guide breaks down the basics of interviewing an expert and creating accurate documentation of their work.

Clarify the goals of the documentation ahead of time

Identify what the documentation should primarily focus on. This will dictate the granularity of your interviews and can have multiple benefits to the readers:

  • Self-reflection or alignment of the team Requires basic granularity of content in order to facilitate an understanding of each person's contributions and value generation.

  • Education and onboarding of new contributors Requires educational tone and higher granularity of content, which may and should be extended in the future.

  • Note-taking and knowledge management Requires primarily scaffolding for content with custom granularity

Then ask the following:

  • How many separate roles can you categorize your work into? If you're not sure, let's stick with one.

  • What are your biggest challenges in the work?

Ideally, refer to the results of the Organizational Mapping [2 hours] workshop to make identifying the right content easier.

Research existing information

  1. Review the existing documentation regarding the issue and goal (GitBook, Notion, etc.)

  2. Get a sense of general familiarization with their ecosystem, products and mission through Google, website, blog, etc

  3. Collect all information into a Notion page for the project. This will make it easier to share and double-check internally.

Conduct interviews to gather insights

Interviews with domain experts to get a clearer sense of the workflows. Please refer to the following guide for a detailed list of questions:

During the interview, make sure you capture insights into the following areas:

  • Goals - What is the function supposed to accomplish? The goals need to be clear and tangible outcomes of executing the function successfully.

  • Hazards - What can go wrong when the function is executed inadequately or not enough attention is paid to it? Ensure capture of the larger-scale, observable consequences of the inadequate operation.

  • Requirements/constraints - What is required to execute the function successfully? What does the operator need in order to do their work well? What are the constraints of the function and what should not be done?

  • Workflow - A granular breakdown of all important steps of the function. It should be educational and act as a guide and reference for future contributors.

Record, transcribe and notate

Record and transcribe interviews (e.g. with Google Meet). Take notes in Miro during the interview(s). Capture as many points as possible, logging their every word if possible. These will be summarised later with GPT to save time in the workflow creation.

Order notes

Due to the setup process prior to the call, your notes should already largely be ordered. Review the Miro for any miscellaneous notes or new workflows that need to be ordered and cluster them into relevant categories for an easier analysis process. Make sure to order workflows in chronological order if applicable and to create bold headlines for any steps of the workflows that do not already have one.

Revisit the existing documentation

Analyze the organization's current documentation to identify the most logical and relevant location for the new workflow, addition or alteration. Make sure it matches and complements the existing stakeholder journeys.

Consider how it relates to existing content and how users might access or reference it during their journey.

Create the first draft

Translate your notes into a concise, easy-to-follow guide. The guide should provide clear instructions and explanations for each step, ensuring that future operators of the function can easily navigate the process and achieve their goals.

Get the intial rough text from GPT:

  • Drag your mouse to select one column individually on the relevant Interview Workshop Miro board. Copying each column individually will ensure all steps of the workflow are in their proper chronological order when they are pasted.

  • Paste the sticky notes into the relevant function's workflow page or overview page.

  • Do this for each workflow.

  • In Notion, change all bold text into Header 2.

  • Copy all text from the Notion page of the workflow you are refining and paste it into chatgpt with the prompt "rewrite:" before the text. This will give you a good starting point to work with.

  • Copy and paste the text from gpt into the notion. If you have made the bold text H2 in notion, it will be in the proper format.

Go through and read each section. Imagine that you are someone coming into the job for the very first time.

  • What information would you need in order to jump directly into the task?

  • What links, documents, or points of contact would you need, what skills or external resources might you need?

  • Are there any learning resources that you might like to explore?

  • Would a screenshot be helpful?

Fill in any necessary context from the information you have from the call and identify clarifying questions that can help make the workflow more granular. Identify any links, screenshots, documents, or points of contact you would need and highlight the relevant text, leaving a comment with what is needed.

Refine and notate the first draft

Go through the workflow step by step and make sure everything flows together nicely.

Always try to occupy the shoes of someone who is trying to use this information to jump blindly into a position. If one task ends up being very comprehensive or there is a specific type of project the function works on that has it's own workflow, make it into its own page.

Optimize the workflow's organization

Rearrange as necessary. Is there anything that should be separated into two different tasks or anything redundant that can be combined with another task?

Identify opportunities for cross-linking

Look for areas of the documentation where there may be existing documentation for specific processes (eg. best practices, meetings), relevant parties, or documents (eg. knowledge bases, templates, etc.). When you identify something you would like to have a link to, place a comment on the relevant sentence or paragraph asking the person who reviews the documentation to provide the link.

Identify opportunities for further granularity

Look for areas of the documentation that may not be clear enough for someone to jump in and start working. For example, if the workflow says "ensure the product goes through quality control" the person reading the documentation will need to know who is responsible for quality control and how to reach them. When you identify something that needs such clarification, place a comment on the relevant sentence or paragraph asking the person who reviews the documentation to provide clarification. Be specific about what you are looking for (eg. "who is responsible and how can they be contacted).

Orient toward the organization's goal

Orient the documentation toward the goal expressed during the intial interview with the organization. For example:

  • Learning - If the team needs new contributors to learn a lot, the breakdown might require higher granularity.

  • Individual progressions - If there are many functions in the organization, it’s important to see the workflow in context to the others to help an individual understand how they up-skill and progress.

  • Hiring - If the organization needs to hire more workers, focus on information that will help them screen the candidates.

  • Scope definition - To communicate internally how deep/complex a workflow is.

  • Task split - The individual tasks should make it easy to split the work between two individuals. One of them will be the leader, the other one a student.

Compare with the recording and add details

Rewatch the recorded interview to ensure all necessary details have been captured in the documentation. Iron out inconsistencies so the workflow elaborates on what you captured during the interview.

Send for review and ask for comments

Send the documentation to the interviewees for review and ask them to place comments to increase granularity. Make sure they are aware of the purpose of the documentation, like educating other team members or having a record of what they would otherwise need to remember in a different place.

Make sure to remind them of the time constraint on providing comments in order to finish the project.

If you are looking for more targeted feedback, you can also conduct an interview with someone who is new to the organization or project, the expert(s) who provided the workflow or process, or a different expert who is also familiar with the workflow or process.

Review interview

If they are not providing you with feedback, you can conduct a follow-up interview instead:

  1. Have them share their screen so you can see where they are looking

  2. Have the expert walk through the GitBook as if they were a new hire and provide commentary on what they notice.

  3. Ask questions throughout about whether anything is missing and if everything included seems helpful.

  4. Afterward, ask for honest feedback and any areas of improvement

  5. Take notes in Miro, which can then be integrated into the second iteration.

Take notes in the existing Miro document. This will allow for easy iteration of the documentation.

Make sure to send them the documentation ahead of time so that they can review it on their own prior to the call where it will be discussed.

Finish the guide

Add the requested details and finish the workflow guide.

  1. Move notes from Miro into GitBook

  2. Integrate any new feedback

  3. Create the final version with any finishing touches

  4. Double-check all links

  5. Hand the result over to the client.

Update your own workflow before the Retrospective

A retrospective workshop will occur after every project. Update your workflow directly upon finishing the project so that it can be referenced during the retrospective.

Regularly review and update the Documentation Writer workflow to keep it current, accurate, and user-friendly. Solicit feedback from users and team members to identify areas for improvement or clarification. Continuously refine your documentation to better serve the needs of your organization and its users.

Last updated