# Rasas portfolio expansion — focused research

Reviewed 11 October 2026. This is desk research, not customer interviews. The two products below are self-initiated demonstrations for fictional businesses with synthetic data. No demand, willingness to pay, client results or market-wide workflow problem is claimed.

| Rank / business type | Problem | Evidence and date | Proposed feature | Dependencies | Success measure | Uncertainty |
|---|---|---|---|---|---|---|
| 1 / service salon | Requests and confirmations can be confused; staff need one view of availability | Novajo service and homepage pages inspected directly in Chrome, 11 Oct 2026: competitor connects scattered messages to reviewable workflows. This is a design precedent, not customer-demand evidence. | LAYN appointment requests; staff confirmation; overlap checks; history; manual reminder queue | Agreed working hours, staff/service assignments; production database and access controls | Demo request survives refresh; conflicts rejected; report equals records | Interview staff about current tools, no-show rate, rescheduling and handoff before a client implementation |
| 2 / clothing retailer | Product and policy answers can drift across staff | Novajo assistant description inspected 11 Oct 2026: knowledge and human review made explicit. Official Responses API reference inspected 11 Oct 2026 documents grounded input and output structure. Neither proves local demand. | SUDA editable bilingual knowledge retrieval, source cards, enquiry review workflow | Approved catalogue and policies, ownership/update process; model and server if generation is later enabled | Policy edit changes evidence; unknown questions stay unanswered; approved task persists | Need interviews on question volume, languages, catalogue update frequency and acceptable review time |
| 3 / both | Mobile usability and Arabic readability | Jordan MoDEE 2024 household ICT survey, current ministry page retrieved 11 Oct 2026: high household smartphone and internet adoption; household data is not a business-needs survey | Arabic/English, responsive controls, no reliance on desktop tables | Arabic copy, keyboard/touch QA | Forms and dashboards work at 390px without horizontal overflow | Actual customer device mix and assistive technology use unknown |
| 4 / retailer | A coherent identity and useful product explanation need to work together | JoPACC 2025 payments report released 29 Mar 2026 establishes a growing digital-payments context; it does not establish ecommerce demand for this fictional label | Original identity, three creative angles, product imagery, motion film, connected catalogue | Real product information and approved accounts for an actual launch | Complete editable assets and working catalogue; prospective launch measures kept separate | Payment choice, fulfilment and audience willingness to buy require validation |
| 5 / retailer | Campaign reporting can mistake interest for sales | Meta Blueprint current objective/conversion course retrieved 11 Oct 2026 connects business goal, objective, conversion location and performance goal | Small proposed creative test, reviewed enquiry metric, UTMs in a future consented provider | Business-owned ad account/Page, payment method, permissions, tracking and privacy decisions | Qualified enquiries and fulfilled orders separately from clicks | Account-specific objective/placement availability and economics not verified; no campaign run |
| 6 / both | Follow-up can be forgotten or sent without consent | Official WhatsApp developer documentation was attempted 11 Oct 2026, but automated access failed; cannot certify current tariff, template rules or account permissions | Visible demo outbox; idempotent manual reminder evaluation; no messages sent | Future: official platform onboarding, consent, templates, server webhooks, credentials, exact current rate card | One queued reminder per confirmed appointment revision; zero external sends | Live integration and operating cost are unverified; do not promise a free or connected WhatsApp system |
| 7 / small businesses | Small teams need practical instructions rather than a tool list | World Bank Jordan financing release, 30 Jun 2026: MSMEs are central to the economy and finance/digital transition remain priorities; no specific training-demand evidence | Needs audit, reusable prompts, source-check exercise and staff review checklist | Business-specific tasks and approved data | Trainee cites current sources, refuses unsupported answers and catches changed policy | Staff skill gaps and training usefulness need observation/interviews |
| 8 / Rasas portfolio | Events must match what the analytics provider supports | Cloudflare Web Analytics FAQ retrieved 11 Oct 2026 states no custom events or UTM support | Retain existing Cloudflare setup; local review-only event log plus explicit unconfigured external adapter | Separately authorized analytics provider and consent policy for real collection | One event per action; no customer fields; no test event transmitted | Production custom collection is not configured |

## Sources and scope

1. [Novajo services](https://novajo.net/services/) and [homepage](https://novajo.net/) — live Chrome inspection 11 Oct 2026. Learning: explain the business situation, deliverable, human approval and starting point; distinguish client/self-owned/demo examples. Original Rasas copy and assets are authored independently. Competitor timelines and outcome language are not adopted.
2. [MoDEE household ICT survey, 2024](https://www.moict.gov.jo/EN//NewsDetails/Ministry_of_Digital_Economy_and_Entrepreneurship_Announces_Results_of_the_2024_Household_ICT_Usage_and_Access_Survey) — underlying data 2024; retrieved 11 Oct 2026; page publication date not supplied. Useful mobile context, not current SME demand.
3. [JoPACC 2025 payments report release](https://www.jopacc.com/media-center/news/jopacc-releases-its-2025-payments-systems-report) — 29 Mar 2026; retrieved 11 Oct 2026. Digital payment adoption context. Payment processing is outside these demos.
4. [World Bank: Jordan private investment/jobs financing](https://www.worldbank.org/en/news/press-release/2026/06/30/world-bank-approves-us-700-million-in-financing-to-boost-private-investment-and-jobs-in-jordan) — 30 Jun 2026; retrieved 11 Oct 2026. Macro context; not a substitute for customer discovery.
5. [Meta Blueprint: objectives and conversion settings](https://agencies.facebookblueprint.com/student/path/253165-selecting-objectives-course) — current official course retrieved 11 Oct 2026; publication date not supplied. Objective choice depends on actual business goal and conversion location.
6. [Meta Blueprint: selecting campaign objectives](https://www.facebookblueprint.com/student/path/219699-selecting-objectives-course) — current official course retrieved 11 Oct 2026; publication date not supplied. Proposed lead test requires verification in the actual account; no invented results.
7. [Cloudflare Web Analytics FAQ](https://developers.cloudflare.com/web-analytics/faq/) — current official docs retrieved 11 Oct 2026. No custom-event/UTM reporting; existing beacon supports route analytics.
8. [OpenAI Responses API](https://developers.openai.com/api/reference/resources/responses/methods/create) — current official API reference retrieved 11 Oct 2026. Server-only adapter uses POST /v1/responses, explicit model configuration and source-only context. No model connection is enabled or paid calls made.
9. [WhatsApp official pricing](https://developers.facebook.com/documentation/business-messaging/whatsapp/pricing/) and [getting started](https://developers.facebook.com/documentation/business-messaging/whatsapp/get-started/) — attempted 11 Oct 2026, inaccessible to research tools. Current rules/costs remain unverified; third-party claims are deliberately not used to promise an integration.

## Build decisions and discovery questions

Two single-browser applications give visitors an end-to-end task to try without needing accounts. Appointments remain requests until staff confirmation; confirmed appointments reserve staff time across the full service duration. Jordan time is displayed explicitly. Retail answers are deterministic retrieval excerpts, labelled accordingly; AI generation is not connected. Local storage is synthetic, resettable and separate between applications. A visitor never needs to supply real personal data.

Opportunity order weighs usefulness, evidence, feasibility and demonstrability equally (1–5): booking 4/2/5/5=16; knowledge desk 4/2/5/5=16; mobile bilingual experience 5/4/5/4=18 (shared requirement); coherent launch/training package 4/2/5/4=15. Workflow scores are assumptions, not measured market findings. Both app choices are the requested briefs refined with confirmation, sources and review safeguards.

Before any client implementation ask: How many enquiries arrive weekly and through which channels? Who updates availability/policies? Which response needs approval? What causes rework? What is the current baseline? What would justify the cost? Observe the workflow with consent, then agree a small measurable pilot. Do not assume automation alone increases sales.