← Back to work
Flagship · Enterprise UX · Search & discovery · AI interaction design

Designing Program Creation to Make Search Work

Search was the visible request. The deeper problem was incomplete upstream data. I redesigned creation around flexible structure, conditional complexity and AI suggestions that never remove human control.

My roleUX Designer — end-to-end workflow redesign
My contributionResearch synthesis, IA, interaction design, testing and AI feature design
CollaborationProduct, engineering and program teams
Evidence status14 sessions reviewed · public outcome metrics restricted
Case-study viewQuick readThe complete decision story in about 90 seconds
01 · Snapshot

Search quality had to be designed upstream.

A six-field form created records, but not the consistent context that discovery, filtering and later operations needed.

ProblemIncomplete program data weakened every downstream search experience.
Evidence14 recorded sessions · about two hours · English and Hindi.
Core decisionControlled metadata where consistency mattered; custom paths where reality varied.
Outcome statusA documented workflow and stronger data model; public post-launch metrics are unavailable.
PrincipleStructure the data, not the user.

Consistency for search should not force specialised programs into the wrong category.

  1. 01

    Hybrid taxonomyControlled terms, searchable multi-select and a reviewed custom path.

  2. 02

    Conditional complexityReveal operational detail only when it becomes relevant.

  3. 03

    Human-controlled AISuggest, explain and recover—never commit automatically.

02 · The real problem

The obvious request was search. The consequential decision was to redesign creation.

The old flow was fast because it captured little: a name, banner, broad focus area, description, visibility and a certificate checkbox. Audience, geography, ownership and certificate rules were handled outside the product.

Product neededConsistent metadata for useful search
Program teams neededLanguage flexible enough for specialist work
Engineering neededPredictable, reusable data
My decisionHybrid taxonomy + progressive disclosure
03 · Evidence that changed my thinking

Three findings changed the product direction.

01 · Language

Broad categories could not describe specialised programs

Direction: keep controlled terms for search, but support multiple selections and a governed custom value.

02 · Real-world scope

Audience and location were not single-value questions

Direction: support mixed professions and local, multi-state, national, global or online reach.

03 · AI trust

AI was useful as a starting point, not an authority

Direction: make suggestions contextual, editable, replaceable, regenerable and optional.

Research credibility: 14 sessions · English + Hindi · ~2h audio · AI-assisted transcription manually checked.

04 · Key structural decision

Free text was too inconsistent. A rigid taxonomy was too narrow. The answer was a hybrid.

Program basics / Focus areas
What will this program cover?

Select all relevant focus areas

Maternal health ×Community care ×

Reconstructed interaction model based on the documented design; confidential production screens are not shown.

05 · Designing the human–AI contract

AI could reduce blank-page effort only if the user stayed in control.

SituationSystem behaviourHuman control
Intent is clear

Suggest structure from visible program context.

Accept or edit.

Intent is incomplete

Ask for context instead of hiding an inference.

Clarify or continue manually.

Suggestion is weak

Offer alternatives and preserve entered context.

Regenerate, replace or skip.

Decision has consequences

Keep output as an unapproved draft.

Review and confirm.

AI is unavailable

Save existing work and keep the workflow open.

Complete the task manually.

06 · The product

A guided setup flow makes the new structure usable.

1

Define

Sector, focus, audience, location, objective and visibility.

2

Create identity

AI-assisted or manual name and banner paths.

3

Set rules

Topics, contacts and conditional certificate detail.

4

Review ownership

Leads, view/edit permissions and final confirmation.

Program workspace / Create
Step 2 of 4

Create the program identity

Start with a suggestion or continue manually.

Based onMaternal healthCommunity health workersBihar
Choose a starting pointAI suggestions are drafts
Interactive reconstruction: select a draft, choose the manual path or apply a suggestion. It demonstrates documented behaviour without exposing confidential screens.
Conditional state

Certificate rules appear only when needed

Eligibility, issuer and accreditation do not burden every program.

Permission state

Visibility and team access stay separate

Discovery, joining, viewing and editing are treated as different decisions.

Review state

Users confirm the complete program before publishing

Generated and manual choices meet in one accountable final review.

BeforeAfter
Basic program recordSearch-ready program context
One broad focus areaSearchable taxonomy + custom path
Manual identity creationOptional AI drafts + manual control
Certificate and access reduced to togglesConditional rules + separate permissions
07 · Impact + influence

What changed in the design—and what remains unverified.

Design outcome

A stronger foundation for discovery

The workflow defines richer, more consistent inputs for search, filtering, registration and future automation.

Product influence

The search request moved upstream

I reframed discovery as a creation and data-model problem, bringing product, engineering and program-team constraints into the same decision.

Evidence boundary

No invented performance claim

Public post-launch metrics are unavailable, so intended benefits are not presented as measured impact.

After launch I would measure: completion and drop-off, metadata completeness, AI accept/edit/skip behaviour, and search zero-result or reformulation rate.

08 · Reflection

Two lessons I would carry forward.

What I learned: enterprise simplification is not always fewer steps; it is putting consequential decisions in the right order.

What I would validate next: whether people understand the taxonomy and AI provenance quickly, and whether the manual fallback feels equally complete.

Optional · Research details

Depth for the interview conversation.

14recorded sessions reviewed
~2hof usability-test audio
2languages
Manualverification against source audio
Research method and evidence integrity

I synthesised spoken questions, hesitation, self-correction and direct feedback from English and Hindi audio. AI-assisted transcription accelerated processing, but I manually checked findings, excluded unclear speech and did not infer visual behaviour from audio-only evidence.

Supporting synthesis themes

Evidence clustered around search taxonomy, audience and location, AI trust and control, and certification operations. Repeated patterns were kept separate from one-off preferences so frequency did not become a substitute for severity.

Other decisions influenced by research

Certification became a conditional workflow with eligibility, issuer and accreditation detail. Visibility, joining, viewing and editing became separate questions instead of one public/private setting.

How AI supported the design process

AI helped with transcription, early pattern surfacing and content exploration. The recordings remained the source of truth; evidence, interpretation and design response were kept distinct.