DCHS Insights CX Human-Centered Design @ DCHS PROTOTYPE no real data
DCHS Insights CX How We Listen

Eight ways to learn from clients' real experiences

The personas and prototype journeys make the value of client-centered evidence visible, but they remain composites built from our assumptions. These eight methods show how DCHS could move that work off the whiteboard and learn with the clients and frontline staff who actually live these experiences. Each method pairs a factual profile with a synthetic sample of the output it could produce.

The planning question

What do we need to learn, from whom, before we choose what to build or change?

If the answer would not change a decision, the research question is not ready yet.

What this carries forward

The April 2025 personas and journey maps gave DCHS a valuable client-centered hypothesis about fragmentation. How We Listen is how we take that gift off the whiteboard: choose a specific journey, program, service, or population and investigate it with the clients and frontline staff who actually live it.

Choose evidence for a decision

Use the personas to identify where real experience matters, then find a feasible way to learn with actual clients and connect that evidence to a decision.

  1. Start with what the personas reveal

    The composite personas make fragmentation concrete and show the value of a client-centered view, but they are still hypotheses rather than evidence from actual clients.

  2. Move from composite to real experience

    Choose a program, population, workflow, or journey moment where clients and frontline staff can confirm, challenge, or deepen the story on the whiteboard.

  3. Connect the uncertainty to a decision

    State what DCHS does not know and what would be done differently if the research answered it.

  4. Inspect the sample output

    Use the synthetic report to judge whether that kind of deliverable would provide evidence at the right depth for the people who need to act.

  5. Assess feasibility and trust

    Consider staff time, client time, money, access, consent, burden, and whether DCHS could act on an uncomfortable result.

  6. Choose the next move

    Run it, adapt it, combine it with another method, or do not use it—and record why.

Choose a method by its output

Each profile explains the method; each synthetic report shows what a commission could actually hand back. Read both before judging whether the method fits the question.

01 Diagnostic

Mobile ethnography

Clients record their own experience as it happens — photos, voice notes, and diary entries on their phone over days or weeks, guided by daily prompts.

Aboul & Hossina

$$$ 3–8 wks 10–20+
02 Generative / synthesis

Customer journey mapping

Staff and clients map one journey end to end in a working session — every step, wait, and emotion — then turn the pain points into owned, prioritized fixes.

Janice Brown

$$ 1 day–6 wks 5–15 in room
03 Monitoring

Rapid high-volume feedback (kiosks)

A one-tap kiosk at the exit gathers hundreds of quick ratings, powering an always-on dashboard that shows where and when the experience dips.

Marisol Reyes

$–$$ Continuous Hundreds+
04 Evaluative / oversight

Quality Service Review (QSR)

Certified reviewers examine one cross-program case in depth, scoring both how the client is doing and how well the system worked together.

Janice Brown

$$$ Weeks–months Dozens of cases
05 Evaluative / legitimacy

Community reviews

Trained community members visit a service, observe the experience, and hear from clients — then publish findings the provider formally answers.

Aboul & Hossina

$$ 4–8 wks Small teams per visit
06 Diagnostic

Shadowing

A trained observer accompanies one consenting client through a real visit, time-stamping every step, wait, and handoff as it happens.

Marisol Reyes

$ Days–2 wks Handful of journeys
07 Diagnostic (staff-side)

Staff friction logs

Frontline staff briefly log each obstacle they hit during normal work; two weeks of entries become a ranked map of what breaks most often.

Janice Brown

$ ~2.5 wks Frontline staff
08 Diagnostic (+monitoring core)

Exit intercepts with depth

Brief structured conversations with clients on their way out — a running tally of fixed questions, plus deeper interviews with a smaller sample.

Janice Brown

$ 1–2 wks+ 12–15/segment depth

Compare the methods

Every method on the same terms: what it can tell us, which layer of listening it belongs to, what it costs to run, and the form of evidence it returns.

Method Says / does Qual / quant Layer Elapsed time Cost Participants Skill needed Primary output
01 Mobile Eth Behavioral Qual Diagnostic 3–8 wks $$$ 10–20+ Med–High Insight report + media reel
02 Journeys Both Qual (+quant overlay) Generative / synthesis 1 day–6 wks $$ 5–15 in room Med Journey map with emotion curve
03 Kiosks Attitudinal Quant Monitoring Continuous $–$$ Hundreds+ Low Live trend dashboard
04 QSR Both Qual + scored Evaluative / oversight Weeks–months $$$ Dozens of cases High / certified Scored findings + improvement plan
05 Community Rev. Both Qual Evaluative / legitimacy 4–8 wks $$ Small teams per visit Med (trained) Published visit report + provider response
06 Shadowing Behavioral Qual Diagnostic Days–2 wks $ Handful of journeys Low–Med Time-stamped flow map
07 Friction Logs Both Qual (tallied) Diagnostic (staff-side) ~2.5 wks $ Frontline staff Low–Med Ranked friction map
08 Intercepts Attitudinal (+reported behavior) Mixed Diagnostic (+monitoring core) 1–2 wks+ $ 12–15/segment depth Med Trend tally + themed depth quotes

$ = low / staff-time-only · $$ = moderate (platform or coordination) · $$$ = high (certified labor). Bands are planning heuristics, not guarantees.

Research dimensions

Every method profile places its method on four dimensions — the first three are the standard user-research axes from Rohrer / Nielsen Norman Group. No single position is best; a listening system needs coverage across the dimensions.

Method Attitudinal ↔ Behavioral what people say vs what they do Qualitative ↔ Quantitative meaning and why vs counts and how many Generative ↔ Evaluative discovering the unknown vs measuring the known Retrospective ↔ In-the-moment recalled after vs captured live
01 Mobile Eth Behavioral Qualitative Generative In-the-moment
02 Journeys Mixed Qualitative Generative Retrospective
03 Kiosks Attitudinal Quantitative Evaluative In-the-moment
04 QSR Mixed Mixed Evaluative Retrospective
05 Community Rev. Mixed Qualitative Evaluative In-the-moment
06 Shadowing Behavioral Qualitative Generative In-the-moment
07 Friction Logs Mixed Qualitative Generative In-the-moment
08 Intercepts Attitudinal Mixed Mixed In-the-moment

Leanings are approximate — orientation, not measurement. "Mixed" means the method genuinely works both sides of the axis.

Questions before commissioning research

A method is only useful when its evidence is needed, feasible to gather, and connected to an action DCHS is prepared to take.

What could this method tell us that we do not already know?

Which people or moments would it systematically miss?

What would it cost in staff time, client time, money, and trust?

Could the relevant team run it well, and what would break first?

If this output arrived next week, what decision or behavior could it change?

What is the smallest ethical pilot that would tell us whether the method is useful here?

What to carry forward

A research target brief

Carry forward a specific target, a consequential uncertainty, the decision it will inform, and a feasible first step.

Program, population, or journey moment What we need to learn Decision it will inform Method or combination Staff or partners needed First step
 

DCHS Insights CX

This planning site is for internal use. Please enter the access password to continue.