Adjacent Product Research Prompt Generator

Produces a structured deep-research brief for evaluating what products to build next alongside the current hero product. Not a brainstorm list — a research framework with evidence requirements, scoring rubric, and MVP roadmap instruction baked in.

When to use

Use after lp-offer (MARKET-06) completes, when the operator intends product range expansion or when GATE-PRODUCT-02-01 fires. Requires offer.user.md to exist for the target business (ICP and pain/promise map are essential inputs that naming research does not need but this skill does). Also operator-invocable at any time after offer exists.

Operating mode

RESEARCH BRIEF AUTHORING ONLY

Allowed: read startup loop docs, synthesize context, produce research prompt artifact.

Not allowed: generate product concepts, make sourcing or pricing decisions, implement any changes.

Required inputs (pre-flight)

Before writing the prompt, read and synthesize from the business's loop docs:

• docs/business-os/startup-baselines/<BIZ>/<YYYY-MM-DD>-assessment-intake-packet.user.md — ICP, product, region, price range, growth intent
• docs/business-os/market-research/<BIZ>/latest.user.md — competitor landscape, pricing benchmarks, manufacturer accessory catalogues
• docs/business-os/strategy/<BIZ>/<YYYY-MM-DD>-brand-identity-dossier.user.md — emotional territory, anti-criteria, community sensitivity framing
• docs/business-os/startup-baselines/<BIZ>/offer.md OR docs/business-os/strategy/<BIZ>/offer.user.md — ICP segmentation, pain/promise map, objection map, positioning (required; this skill cannot run without it)
• docs/business-os/contracts/<BIZ>/outcome-contract.user.md — 90-day targets and contract window (to anchor MVP roadmap dates)

If the offer artifact is missing, halt and instruct the operator to complete lp-offer (MARKET-06) first.

If any other file is missing, note the gap in the output but proceed with available context.

Output

Two artifacts:

Prompt (produced by this skill):

docs/business-os/strategy/<BIZ>/lp-other-products-prompt.md

Results (produced by human after running in a deep research tool):

docs/business-os/strategy/<BIZ>/lp-other-products-results.user.md

Note the results path in the completion message. The human saves the tool's output there. lp-prioritize (post-S4) picks up product candidates from Track 4's top-5 shortlist as additional go-items when that file is present.

Prompt structure (mandatory sections)

The generated prompt must contain all of the following sections in this order:

1. Researcher role and output rules

Instruct the researcher to:

• Produce a structured product range strategy document, not a generic brainstorm
• Use web research and cite sources for all factual statements (competitor offerings, manufacturer catalogues, price points, materials, regulatory definitions)
• Separate clearly: (a) what was checked (with source + date), (b) what was inferred, (c) what remains unknown without supplier quotes, lab testing, or legal counsel
• Make no medical efficacy claims; propose no products whose core promise is to improve hearing outcomes

2. Date, locale, and currency

Encode from intake packet:

• Today's date (Europe/Rome timezone for EU-primary businesses)
• Launch market (primary and expansion)
• Currency focus (EUR primary, USD note if relevant)

3. Business context

Cover at minimum:

• Current hero product (physical description, function, what it does for the user)
• Positioning statement (lifestyle accessory vs medical device; no medical device claims)
• Primary customer (ICP, from offer artifact — include JTBD and buying trigger)
• Secondary customers
• Current price range and core tier
• Emotional territory — encode the 5 brand personality descriptors from brand dossier
• Market sizing — label as verify-and-cite; instruct researcher to verify figures and cite with dates

4. Regulatory and claim-drift guardrails

This section is mandatory for health-adjacent products. Encode:

• Cite EU MDR definitions of "intended purpose" and "accessory for a medical device" — instruct researcher to cite the relevant MDR passages, not just describe them by feel
• Require a claim drift risk rating (low / medium / high with rationale) for every adjacent product category proposed
• Permitted benefit language: comfort, secure wear, reduced drops/loss, everyday usability, convenience, organisation, personalisation
• Prohibited framing: restore hearing, improve hearing, therapy, treatment, necessary for the device to function

5. Scope definition

Define what "adjacent" means for this brand:

• Products used with or around the external processor / wearing routine / daily life of users and families
• Expand basket size and repeat purchases (sizes, colours, seasonal drops, replacements, add-ons)
• Feasible as a small DTC brand (manufacturing, compliance, shipping, returns)
• Extensible into a coherent range architecture (not one-off oddities)

6. Research tracks

Always include all six tracks:

Track 1 — Customer JTBD and moments map (evidence-led): Map the user journey across: morning routine / school / play / sport; sleep / bedtime; travel / beach / rain / sweat; storage, organisation, cleaning, "where did it go?" Identify the top 10 recurring problems that adjacent products could solve. Require citations from community discussions, manufacturer guidance, or reputable resources.

Track 2 — Competitor and manufacturer accessory landscape: Build a structured map with two separate classes: (A) independent accessory brands, (B) manufacturer official accessory catalogues (fixation, clips, covers, cases, drying/cleaning, sports solutions). For each: product types, naming/positioning style (functional vs lifestyle), typical price bands (cited), notable gaps. Cite primary sources.

Track 3 — Adjacent product candidate set (range blueprint): Generate 12–20 product concepts grouped into a range architecture (e.g. core extensions, add-ons, retention ecosystem, storage/organisation, care/maintenance, parent convenience). For each concept, require a one-page card with these 11 fields:

1. What it is (plain description)
2. Primary customer + use moment
3. Value proposition (non-medical framing)
4. Expected ASP range (EUR) with citations
5. Suggested materials / form factor and why
6. Size/fit complexity (low/med/high)
7. Manufacturing complexity (low/med/high)
8. Returns risk (low/med/high)
9. Compliance/claim-drift risk (low/med/high) with rationale
10. Competitive references (who sells something similar; cite)
11. Range role (entry add-on / premium / replenishment / bundle anchor)

Track 4 — Prioritisation and MVP roadmap: Require a scoring rubric with explicit weights, e.g.:

• Customer impact (30%)
• Differentiation / whitespace (20%)
• Manufacturability + supply chain simplicity (20%)
• Margin potential at target ASP (15%)
• Claim-drift / compliance risk (15%; inverted — higher risk = lower score)

Output:

• Top 5 adjacent products to explore next (with rubric scores)
• Top 3 to build first with a 90-day MVP plan: prototyping approach, sampling, minimum viable variants (SKUs), basic QA, photography needs, launch bundle strategy
• 12-month range roadmap: how the catalogue expands without becoming incoherent

Track 5 — Bundling, pricing, and merchandising strategy: Propose: bundle archetypes (starter kit, school kit, sport kit, sleep kit, travel kit); cross-sell logic on PDP and cart; replenishment / replacement triggers (growth, lost items, wear-and-tear). Keep price points consistent with the base product range.

Track 6 — Brand and community sensitivity check: For each adjacent category, flag: concealment framing risk (avoid); disability-identity misstep risk (flag if not validated); language guidelines with copy do/don't examples.

7. Deliverables format

Require the researcher to produce a research document (en-GB) with:

1. Executive summary — range thesis and top opportunities
2. JTBD map — top 10 problems with evidence
3. Competitive and manufacturer accessory landscape — structured table with citations
4. Range blueprint — 12–20 product cards grouped by range role
5. Prioritised shortlist — scoring table with weights shown
6. 90-day MVP plan — top 3 products
7. 12-month range roadmap and bundling strategy
8. Risk register — claim-drift/compliance risks and operational risks

8. Reminders

Always end the prompt with:

• No medical claims; avoid implying improved hearing
• Cite everything factual
• Clearly separate checked vs inferred vs unknown
• Use Italy-first lens (language accessibility, pricing expectations, channels)

Quality gate

• Offer artifact read and ICP/pain-map encoded in business context section
• EU MDR claim-drift guardrails encoded with low/medium/high rating instruction per product category
• JTBD map across all 6 user moments requested (Track 1)
• Manufacturer official accessory catalogues listed as a separate comparator class from independent brands (Track 2)
• Product card format with all 11 fields specified (Track 3)
• Scoring rubric with percentage weights present in prioritisation (Track 4)
• 90-day MVP plan and 12-month range roadmap explicitly requested (Track 4)
• Bundle archetypes and replenishment triggers requested (Track 5)
• Community sensitivity check covering both concealment risk and disability-identity risk (Track 6)
• Checked/inferred/unknown separation required throughout
• Market sizing flagged as verify-and-cite

Completion message

"Product range research prompt ready: docs/business-os/strategy/<BIZ>/lp-other-products-prompt.md. Drop into a deep research tool (OpenAI Deep Research or equivalent). When results are returned, save to docs/business-os/strategy/<BIZ>/lp-other-products-results.user.md — lp-prioritize (post-S4) will pick up product candidates from Track 4's top-5 shortlist as additional go-items when that file is present."