From Stories to Hypotheses: A Participatory Framework for Consulting Discovery

A consulting engagement often begins with a familiar pattern.

A sponsor describes a problem. The consultant interviews a small number of leaders. The consultant reviews a few documents, identifies themes, and returns with a diagnosis, a recommendation, or a proposed intervention.

There is nothing inherently wrong with this pattern. A sponsor’s perspective matters. Interviews matter. Documents matter. Consultants need to form a point of view.

But this approach has a limitation: the people who are easiest to reach at the beginning of an engagement are not always the people who encounter the situation most directly.

The consultant may hear a coherent account of “why the change is stalled” from a sponsor and two senior leaders. Meanwhile, people responsible for implementation may be navigating a different reality: unclear decisions, conflicting requests, workarounds that never reach leadership, customer consequences, or practical constraints that are difficult to express in an executive meeting.

The answer is not to treat leaders as unreliable or to collect every possible opinion.

It is to create a more deliberate loop:

1. Listen to a wider range of people.
2. Invite accounts of concrete experience, not only abstract feedback.
3. Invite participants to reflect on what their own experiences mean.
4. Look for patterns across those interpretations.
5. Develop tentative hypotheses about what those patterns may suggest.
6. Return to the associated stories, including exceptions and counter-narratives.
7. Ask the client what the interpretation gets right, misses, or needs to explore further.
8. Use that conversation to decide what should happen next.

This is not a way to automate organizational diagnosis.

It is a way to make consulting discovery less dependent on a small number of initial accounts—and less likely to treat the consultant’s first interpretation as the final word.

The limits of the usual stakeholder sample

Most consultants know that a client request is only a starting point.

A sponsor may say:

“The programme has lost momentum.”

A functional leader may say:

“Teams are not taking ownership.”

A project lead may say:

“We need better alignment.”

These statements may all be sincere. They may also be incomplete.

A small set of interviews can give a consultant depth, context, and access to decision-makers. But it can also create an accidental narrowing of the inquiry.

The people selected for interviews may:

• Hold similar positions in the hierarchy
• Share the same language for describing the problem
• Have strong incentives to frame the situation in a particular way
• Be removed from the day-to-day consequences of decisions
• Be too busy to describe the practical detail that would change the consultant’s understanding
• Be unaware of workarounds, tensions, or risks experienced elsewhere in the organization

This does not make senior input less valid. It means that a discovery process may need more than a few authoritative perspectives.

The question is not:

How can I collect more data?

The better question is:

Whose experience would change, complicate, or qualify my current understanding of the situation?

Listen for experiences, not only opinions

When consultants invite a broader group to contribute, the inquiry often becomes a survey.

Participants are asked to rate statements:

• How clear are the priorities?
• How aligned is your team?
• How confident are you in the change?
• What are the three biggest obstacles?

These questions can be useful for a directional signal. But they often produce numbers, rankings, and lists without showing the situations that give those answers meaning.

Compare these two responses.

Question: What are the biggest barriers to delivery?

Unclear priorities. Slow decisions. Lack of capacity.

This may be accurate. It is also difficult to act on.

Now consider an experience-focused invitation.

Question: Think of a recent occasion when delivery became more difficult than expected. What were you trying to accomplish? What happened? What made the situation difficult?

We were asked to commit to a launch date before the technical dependencies had been agreed. We raised the dependency twice, but no team had authority to resolve it. When the date slipped, the delivery team was told that it had not coordinated effectively.

The second response does not prove a diagnosis. But it gives the consultant more material to work with.

It raises questions about decision rights, planning assumptions, escalation paths, accountability, and how different groups define coordination.

A useful discovery inquiry asks people to share a situation they encountered rather than asking them to summarize the organization.

A reusable experience prompt

The prompt does not force someone to write a long story. It gives enough structure for them to move from a generalized view toward an account of something that occurred.

Gather enough perspectives to see variation

A consultant does not need hundreds of responses to benefit from wider listening.

Even a modest inquiry can broaden discovery when it includes people who encounter the work from different positions:

• Sponsors and senior decision-makers
• Functional or team leaders
• Programme and project leads
• Frontline managers
• Individual contributors
• Operations, technical, customer-facing, or support roles
• People affected by a decision but not normally included in planning conversations

The purpose is not statistical representation. It is not a referendum. It is not an attempt to make every perspective equally decisive.

It is to understand variation.

A strategic initiative may look urgent from the executive team’s perspective, fragile from the implementation team’s perspective, and confusing from the perspective of people closest to customers. Each account may make visible conditions that the others cannot see.

Cerne.io gives the consultant a practical way to run a wider inquiry without losing the connection between an account of experience and the participant’s own interpretation of it.

Instead of collecting narrative input through a generic form, exporting it into a spreadsheet, copying fragments into slides, and manually trying to remember who was describing what, the consultant can keep stories, reflective responses, hypotheses, excerpts, and client-facing materials connected within one project.

This makes it easier to:

• Invite a relevant group to share situated experiences
• Ask participants to reflect on what their own experience means through carefully designed prompts or sliders
• Review patterns across those reflections before deciding which hypotheses are worth exploring
• Filter the stories associated with a pattern by relevant participant context, such as role, function, location, or tenure
• Test whether a pattern appears broadly or is concentrated in one part of the organization
• Return from an interpretive pattern to the full stories, rather than treating a visual cluster as self-explanatory
• Preserve a traceable path from participant interpretation to story, hypothesis, qualifying account, and client-facing question

The goal is not to replace interviews. A focused experience-sharing inquiry can complement them, while adding something that conventional interview-led discovery often misses: participants can interpret their own experiences through carefully designed reflective prompts.

The consultant can then look first for patterns in those interpretations, develop tentative hypotheses, and return to the associated stories and follow-up conversations to understand what the patterns may mean in context.

The consultant remains responsible for deciding what level of breadth and depth is appropriate.

1. Start with how participants interpret their own experiences

A story tells a consultant what happened from one person’s point of view.

But a story does not arrive with a single, self-evident meaning.

The same account of a delayed decision, a difficult handoff, or a successful workaround can be understood in different ways. One participant may experience a delay as evidence of careful risk management. Another may experience a similar delay as unnecessary bureaucracy. A consultant may be tempted to interpret both accounts as “poor communication,” even though neither participant would describe the situation in that way.

For this reason, discovery does not need to rely only on a consultant reading stories and assigning themes afterwards.

After sharing an experience, participants can also be invited to reflect on what that experience means to them.

In Cerne, this can happen through reflective prompts or sliders that ask a participant to locate their experience between meaningful possibilities.

For example:

• Did this situation feel more like a local exception or a recurring pattern?
• Did it arise more from unclear decisions or from practical constraints?
• Did it feel more like a productive trade-off or an avoidable friction?
• Did the people involved have enough room to act, or were they constrained by decisions made elsewhere?
• Did this experience increase confidence in the work, or make progress feel less possible?

The specific prompts must be designed for the purpose of the inquiry. They are not diagnostic instruments, and they do not reveal an objective truth about an organization.

Their purpose is to capture how participants themselves make sense of the experience they have chosen to share.

This approach draws on Participatory Narrative Inquiry, which uses stories of lived experience to support collective sensemaking in complex situations.

This creates a different starting point for analysis. Instead of beginning with a consultant’s coding of story text, the consultant can first examine patterns in the meanings participants have assigned to their own experiences.

For example, a cluster of stories may be interpreted by participants as:

• Recurring rather than exceptional
• Constrained by decisions made elsewhere
• More frustrating than productive
• Associated with low confidence that concerns will lead to action

That pattern does not explain itself. It gives the consultant a focused place to look, a question to bring back to the stories, and a basis for a tentative hypothesis.

Cerne keeps each participant’s interpretation connected to the experience they shared. The consultant can see an emerging pattern in the reflective responses, then open the associated accounts to understand what kinds of situations, relationships, and conditions may be contributing to it.

2. Form hypotheses from interpretive patterns, then return to the stories

Patterns in participant interpretations can suggest useful working hypotheses.

A working hypothesis is not a conclusion. It is a provisional interpretation that helps the consultant and client decide what to investigate, discuss, or test next.

For example, imagine that many participants locate their stories toward:

• Decisions made elsewhere rather than decisions they could influence
• Recurring rather than isolated situations
• High effort with low confidence that the effort will change anything

This pattern might lead to a hypothesis such as:

It may be difficult for teams to make progress because important decisions are experienced as distant from the work, while the people responsible for delivery carry the consequences.

The hypothesis does not come from a consultant selecting a few striking quotes and declaring a diagnosis. It begins with a pattern in how participants interpreted their own experiences.

The next step is to return to the stories.

Read the accounts associated with the pattern and ask:

• What situations are participants describing?
• Which decisions feel distant, and to whom?
• What practical consequences are people encountering?
• Are different kinds of stories sitting in the same interpretive area?
• What conditions make the pattern more or less likely?
• Which accounts qualify or challenge the emerging hypothesis?
• Where does the hypothesis not apply?

This is where stories matter most. They provide the context that a slider, cluster, or visual pattern cannot provide on its own.

A consultant may find accounts such as:

“We were asked to commit to a launch date before the technical dependencies had been agreed. We raised the dependency twice, but no team had authority to resolve it.”

“Everyone says we own the outcome, but no one can make the dependency decision.”

But the consultant may also find an important counter-example:

“For our team, dependencies are usually clear. The difficulty starts when commercial commitments change after planning.”

The counter-example does not invalidate the pattern. It helps the consultant understand its boundaries.

Perhaps the issue is not decision distance everywhere. Perhaps it appears at a specific handoff, in a particular type of project, or when commitments change after planning.

This is why Cerne keeps interpretive patterns, original stories, selected excerpts, and working hypotheses connected. A consultant can move from a pattern in participant meaning-making to the accounts that give it context, then prepare a client-facing hypothesis that includes both support and qualification.

A good working hypothesis should be:

• Grounded in a visible pattern in participant interpretations
• Connected to the stories that help explain the pattern
• Written in conditional language
• Specific enough to be discussed or tested
• Open to correction
• Accompanied by relevant exceptions or counter-narratives
• Useful for choosing a next step

Avoid statements that turn interpretation into fact:

The disciplined use of conditional language matters. It prevents the consultant from presenting an early interpretation as settled fact—and makes it easier for the client to challenge, qualify, or improve the hypothesis.

Use stories and excerpts to explore a pattern, not prove a conclusion

A pattern in participant interpretations can point the consultant toward an important question. It cannot explain itself.

That is why the consultant returns to the stories.

Selected excerpts help a client understand what participants were describing when they interpreted their experiences in a particular way. They preserve the situations, relationships, decisions, and consequences that are invisible in an aggregate pattern.

For example, a cluster of stories may have been interpreted by participants as recurring, constrained by decisions made elsewhere, and associated with low confidence that concerns will lead to action.

The consultant might develop this working hypothesis:

Teams may be asked to carry delivery responsibility without sufficient influence over the decisions and dependencies that shape delivery.

The consultant can then select excerpts that help the client explore—not “prove”—that possibility.

“We agree to dates before we know whether the integration team can support the work.”

“Everyone says we own the outcome, but no one can make the dependency decision.”

At this point, the consultant should actively look for counter-evidence.

Perhaps another participant says:

“For our team, the dependencies are clear. The delays usually begin when commercial commitments change after planning.”

That counter-narrative does not make the working hypothesis useless. It helps define its limits.

The emerging pattern may apply primarily to a particular type of work, a specific handoff, a certain stage of planning, or a subset of teams. It may point to a different hypothesis altogether.

A useful client readout can therefore include:

This approach is more credible than selecting only quotes that support the consultant’s preferred explanation.

It also helps the client distinguish between:

• A pattern worth exploring
• An isolated incident
• A perspective that needs more context
• A meaningful exception
• A disagreement about what is happening
• A hypothesis that needs revision

Cerne can make this workflow more manageable by keeping excerpts attached to a working hypothesis, including both relevant and qualifying accounts, and allowing the consultant to return to the full original response when more context is needed.

Instead of maintaining one version of quotes in a spreadsheet, another in presentation slides, and another in a working document, the consultant can prepare an inspectable hypothesis view where the client can see the interpretive pattern, provisional hypothesis, selected stories, counter-evidence, and open questions together.

Use excerpts carefully:

• Remove names and identifying details unless explicit permission has been given
• Do not use a quote as evidence against an individual or team
• Avoid selecting only comments that support a preferred conclusion
• Include qualifying accounts or counter-examples where they materially affect the interpretation
• Preserve uncertainty when accounts conflict
• Do not share material in a way that makes participants identifiable in a small group
• Make access and confidentiality arrangements clear before collecting input

A quote is not a verdict. It is a window into an experience that deserves careful interpretation.

Return to the client for sensemaking

The most important step comes after this initial exploration.

Do not treat the consultant’s analysis as a finished diagnosis delivered to the client. Bring the emerging hypotheses back into a structured conversation.

In participatory inquiry, this kind of return step is sometimes called member checking: preliminary interpretations are shared with people who have relevant knowledge of the situation so they can correct, qualify, challenge, or extend them.

In consulting, the aim is not a formal research-validation exercise. It is to avoid letting the consultant’s first analysis become the final story.

This can be a sponsor discussion, a leadership working session, a cross-functional sensemaking session, or a smaller group chosen for the purpose.

The question is not:

Do you agree with my findings?

That invites defensiveness and a binary response.

Instead ask:

• What in this hypothesis resonates with your experience?
• What feels incomplete, overstated, or mistaken?
• What context is missing?
• Which excerpts point to a pattern worth exploring further?
• Which excerpts suggest an important exception or disagreement?
• Where do different groups understand the situation differently?
• What would we need to learn before deciding whether this hypothesis is useful?
• Which question matters most to address next?

Cerne can support this moment with a client-facing sensemaking view: an inspectable presentation of the interpretive pattern, the provisional hypothesis, selected non-attributed excerpts, qualifying accounts, uncertainty, and the question the client is being asked to consider.

Rather than receiving a slide titled “Key Findings,” the sponsor can work with a view that makes the consultant’s reasoning visible:

This changes the relationship between consultant and client.

The consultant is still bringing expertise, interpretation, and process design. But the client is invited to contribute further context and judgment, rather than simply accept or reject a finished diagnosis.

A good sensemaking session does not need to force agreement.

Its outcome may be:

• A revised hypothesis
• A more precise question
• A decision to gather further input
• A clear disagreement that needs to be addressed openly
• A narrower, more realistic intervention
• A decision to test a change in practice
• A recognition that a proposed workshop is not yet the right next step

That is progress. The purpose is better collective judgment, not premature consensus.

A practical client sensemaking format

A 60–90 minute sensemaking session can follow this structure.

1. Revisit the purpose

State why the inquiry was conducted and what it can reasonably show.

For example:

“We invited people to share recent experiences related to this work and reflect on what those experiences meant to them. The purpose was to broaden our understanding before deciding what intervention would be most useful. These materials are not a complete diagnosis, and they do not represent every perspective.”

2. Share a small number of hypotheses

Present no more than three to five provisional hypotheses.

For each one, include:

• The interpretive pattern that prompted the inquiry
• The hypothesis in conditional language
• Two or three non-attributed stories or excerpts that help explore it
• One qualifying account, counter-example, or area of uncertainty where relevant
• The question the hypothesis raises
• The practical implication if the group decides it is worth exploring

For example:

3. Invite interpretation before solution-making

Give the group time to consider:

• What is familiar here?
• What is surprising?
• What needs qualification?
• What is absent?
• What would be risky to assume?
• What is the group not yet ready to conclude?

Do not move immediately into solution brainstorming. The purpose of this stage is to improve the group’s account of the situation.

4. Decide the next useful move

Close by identifying one or two next steps:

• A focused workshop question
• A follow-up inquiry
• A decision that needs clarification
• A change to a process or governance mechanism
• A direct conversation between roles or teams
• A test or experiment
• A revised engagement scope

The client should leave with a clearer question and a more grounded next action—not merely a slide deck of themes.

A workflow for participatory discovery

Cerne supports a consulting discovery workflow that moves from participant experience to client sensemaking without prematurely separating stories, participant interpretation, and consultant hypotheses.

1. Define the initial client question.

2. Launch a focused participant collection space.

3. Invite a relevant group to share recent experiences.

4. Invite participants to reflect on the experience they shared through carefully designed prompts or sliders.

5. Review patterns in how participants interpreted their own experiences.

6. Open the associated stories and compare them across roles, functions, locations, tenure, or other relevant participant context.

7. Develop a small number of provisional hypotheses, then attach relevant stories, qualifying accounts, counter-narratives, uncertainties, and questions to each hypothesis.

8. Create an inspectable client sensemaking view.

9. Bring the material back to the sponsor or client group for correction, qualification, interpretation, and choice.

10. Record what the client learns and what the engagement should do next.

This workflow reduces the mechanical work that can otherwise distort consulting discovery.

Instead of copying fragments between surveys, spreadsheets, notes, and slides, the consultant can keep the inquiry, participant interpretations, stories, filters, hypotheses, excerpts, and client-facing readout connected in one project.

That does not replace consultant judgment. It gives the consultant more time and a better structure for using that judgment carefully.

Consulting discovery is a conversation, not a harvest

Clients hire consultants partly because they want an outside perspective, a method, and someone who can make sense of complexity.

That outside perspective is strongest when it remains connected to the people living with the situation.

Listen beyond the first few voices. Invite accounts of experience. Ask people to reflect on what those experiences mean to them. Begin with patterns across those interpretations, then return to the stories to understand, qualify, and challenge what the patterns may suggest.

Develop hypotheses carefully. Keep them connected to both relevant and dissenting accounts. Return them to the client as material for further sensemaking.

The result is not a faster diagnosis.

It is a more grounded consulting conversation—and a better basis for deciding what to do next.