Method — IT Architecture Options

A multi-option architecture evaluation frame. Given a business target state, compare three or more architectural paths to get there: enhance-legacy, best-of-breed-package, integrated-package (or equivalents per context). The output is a pptx deck with a side-by-side summary matrix that makes the decision legible to a board.

GrowDirect ships this as the consulting:it-architecture-options skill. Plug in a retailer's current application landscape and target operating model; get a structured multi-option deck with per-option heat-maps, effort estimates, risk, and a side-by-side comparison.

The eight-section frame

1. Introduction — purpose, scope, options under evaluation.
2. Programme Targets — margin / overhead / headcount / sales / profit targets the architecture must enable.
3. Business Requirements — structured by process frame (Sell / Plan / Move / Buy or equivalent). This section is domain-driven, not technology-driven.
4. Implications for Legacy Systems — what the current landscape can and can't deliver against the target state.
5. Option A — Legacy Enhancement (7-slide skeleton below)
6. Option B — Best-of-Breed Package
7. Option C — Integrated Package
8. Summary and Conclusion — side-by-side comparison, recommendation.

Per-option slide skeleton

Every option gets the same seven slides, in the same order. The symmetry is what makes the decision legible.

1. Application architecture heat-map — target vs. optimisation processes, 4-color legend (full support / partial / gap / not applicable).
2. Aspirational heat-map — same structure, against future-state processes.
3. Development effort estimate — design days, build-test days, rollout days, converted to man-years.
4. Target application architecture — post-change view of the landscape.
5. Advantages / Disadvantages — balanced list, not a pitch.
6. Implementation timeline and risk — phased plan with named milestones; risk register.
7. Resource / sourcing plan — who builds it (internal, vendor, SI partner), with FTE and skills breakdown.

The decision matrix (section 8)

The side-by-side summary is a single slide that lets a board make the call. Columns are the options; rows are the decision dimensions.

The recommended column is filled last, after the evidence in the preceding rows has been established.

What makes this evaluation durable

1. Options are symmetric. Every option gets the same seven slides. The reader is trained once and compares cleanly after.
2. Heat-maps are honest. The 4-color legend doesn't distinguish between "full" and "excellent"; it distinguishes full / partial / gap / not-applicable. A vendor deck calibrated this way is not a sales pitch.
3. Effort is quantified. Man-years, not "significant effort." Timeline in months, not "18-24 months."
4. Risk is enumerated. Named risks with severity and mitigation, not "there is risk inherent to this path."
5. The recommendation is anchored. Evidence in rows 1–8 of the decision matrix drives the recommendation in row 9. If the evidence doesn't pick the option, the evidence is incomplete.

How the skill applies the frame

The consulting:it-architecture-options skill plugs into:

• Current-state inventory ingest. Application landscape, integration map, current vendor contracts, known gaps. Produces structured input for the heat-maps.
• Option generator. Given the target state and the current landscape, scaffold the three (or more) options with their architectural shape.
• Heat-map renderer. Given an option's target architecture and the business-requirement frame, render the 4-color heat-maps.
• Effort estimator. Given the option's scope of change, produce the design / build-test / rollout / man-years table.
• Decision matrix assembler. Final side-by-side summary.
• pptx output. Deck with the 8-section structure and 7-slide per-option skeleton.

Related

• method/overview
• method/retail-diagnostic
• method/artifacts/sdd-template
• method/artifacts/traceability-matrix-template