eac-ecommerce-bds-programme-design

name	eac-ecommerce-bds-programme-design
description	Use when designing a donor-funded EAC e-commerce business-development-services programme, company-selection method, needs-assessment-to-technical-assistance pipeline, expert-pool deployment model, or results-based monitoring approach.
metadata	{"portable":true,"compatible_with":["claude-code","codex"]}

EAC E-Commerce BDS Programme Design

Acknowledgement: Shared by Peter Bamuhigire, techguypeter.com, [phone].

Overview

Use this skill to design a credible e-commerce business development services programme for East African companies. It joins enterprise support, digital trade, company selection, diagnostics, technical assistance, expert-pool management, monitoring, and learning into one delivery model.

The skill is optimised for the GIZ EU-EAC DEEP style of assignment: select at least ten companies, assess needs, co-create tailored technical assistance for at least five, manage a part-time expert pool, document lessons, and support cross-border digital trade.

Use When

• Drafting a technical methodology for donor-funded BDS, accelerator, enterprise-support, e-commerce, or cross-border digital trade work.
• Designing company selection, diagnostics, TA packages, expert-pool deployment, engagement retention, or monitoring.
• Translating a ToR into a delivery architecture that must be credible under tight page limits.

Do Not Use When

• The assignment is only a marketing campaign, standalone training, or software implementation.
• You do not have the ToR objectives, phases, team model, and selection constraints.
• You need only a social-media call for applications; use the campaign skill instead.

Required Inputs

Artefact	Source	Required?	If absent
ToR objectives, phases, outputs, grid, eligibility, and geography	Procurement pack	required	Stop programme design and identify missing tender evidence.
Fee-day budget, expert roles, company data, and policy evidence	Price schedule, team plan, verified sources	conditional	Mark capacity and selection assumptions provisional.

Workflow

Stop or block the workflow when a required input, permission, or acceptance basis is missing. Recover by revising the scope, obtaining evidence, or returning the narrowest qualified draft before proceeding.

1. Write the theory of change as a results chain: inputs, activities, outputs, outcomes, impact, assumptions, and monitoring questions.
2. Separate eligibility criteria from award criteria. Eligibility screens pass/fail compliance; award criteria score innovation, scalability, cross-border potential, traction, readiness, inclusion, and geographic balance.
3. Design the needs-assessment-to-TA pipeline. Standardise diagnostic domains, then let company evidence decide the TA package.
4. Reconcile expert-pool deployment to person-days. Allocate roles, cadence, backstopping, quality review, and engagement tactics realistically.
5. Build the results framework. Each indicator must have a source, frequency, disaggregation, validation rule, and owner.
6. Create the learning mechanism. Convert company evidence into anonymised case notes, reusable training material, and good-practice synthesis without exposing confidential data.
7. Compress for the proposal. Only include claims that map to a scored criterion or prove delivery feasibility.

Quality Standards

• Every causal arrow in the theory of change has an assumption and monitoring question.
• Selection distinguishes eligibility from award scoring.
• Diagnostics lead directly to tailored assistance, not generic training menus.
• Expert-pool plans reconcile to available fee-days and calendar reality.
• Monitoring indicators derive from the results chain.
• Learning outputs protect company confidentiality.

Anti-Patterns

• Treating a generic capacity-building narrative as programme design. Fix: connect business viability, diagnostics, tailored assistance, and results.

• Selecting firms by size alone. Fix: apply published eligibility and evidence-based award criteria.

• Offering a standard TA menu. Fix: derive each package from the company diagnostic and unit economics.

• Planning more expert support than funded. Fix: reconcile role days, cadence, and backstopping to the fee-day envelope.

• Naming indicators without evidence rules. Fix: state source, validation, disaggregation, frequency, and owner.

• A generic capacity-building narrative with no business-model viability logic.

• Selecting firms by size alone instead of capability, readiness, and growth potential.

• A TA menu disconnected from diagnostics and unit economics.

• Expert-pool claims that cannot fit the fee-day budget.

• Monitoring indicators with no source or validation rule.

Outputs

Artefact	Consumer	Acceptance condition
Programme design pack	Evaluator, programme manager, expert-pool lead	Includes theory of change, selection, diagnostics-to-TA, fee-day deployment, results, confidentiality, learning, and scaling.

Evidence Produced

Evidence	Consumer	Acceptance condition
Selection and fee-day reconciliation	Scoring matrix and resource table	Eligibility is pass/fail, award scores are evidenced, and planned expert days do not exceed the commercial envelope.

Capability and Permission Boundaries

Read and search are required; any edit or external action remains within the explicit authority and permission boundary stated below.

Planning defaults to read-only. Company contact, selection, scoring publication, access to commercial data, expert appointment, funding decisions, and programme launch require explicit authority and conflict controls.

Degraded Mode

Without the tender grid, fee-day budget, company evidence, or verified policy sources, return a provisional programme architecture and evidence requests. Do not name selected firms or claim feasibility.

Decision Rules

Choice	Action	Risk avoided
Mandatory eligibility criterion	Apply pass/fail before scoring	Ineligible award
Needs vary materially by company	Tailor TA from diagnostic evidence	Generic training menu
Expert demand exceeds fee-days	Reduce scope, change cadence, or escalate	Unfunded delivery promise

Worked Example

Screen ten firms for eligibility, score eligible firms against published criteria, diagnose the selected cohort, assign tailored TA to at least five within reconciled expert days, and measure verified business changes without exposing confidential records.

References

• Proposal skills router for repository-wide routing and mandatory quality gates.

• references/programme-design-framework.md: Results chain, selection, diagnostics, TA packages, and learning model.

• references/expert-pool-and-scoring-model.md: Fee-day reconciliation, expert deployment, and proposal scoring discipline.