Custom Software & Legacy Modernization Playbook A bounded first case around a system that already runs, and a first case that can travel. Six families of work, fourteen design tests, and a path that funds an interface or a determined path before a program to replace the core, or an unsupervised overlay, is treated as a first-day design. Not a twelve-week guarantee. Edition: November 2026. 28 pages. Written for: Technology, operations and risk leaders, and the sponsors who have been asked to fund a modernization program or a custom application around a core that already runs. Canonical URL: https://asymilinkmeta.com/resources/custom-software-playbook Written for the search query: custom software legacy modernization playbook Key takeaways - This is the consulting document behind the Custom Software & Legacy Modernization page — not a survey, not a twelve-week guarantee, and not a program plan to replace a core that already runs. - Six families of work. An honest first case is a bounded interface to a system that already runs, or a determined path the existing tools cannot express. A program to replace the core, or an unsupervised overlay, is not a first-day design. - A write to the core is a different design from a read. Build, buy, or leave as-is is asked before a custom path is funded, pointing at the vendor-evaluation scorecard. - Return is measured on a named process against a signed baseline, using the ROI guide. This playbook will not invent a deadline or claim a certified connector to a named platform. What's in this playbook This playbook is the operational document behind AsymiLink Meta's Custom Software & Legacy Modernization consulting page. It is a method for designing a first custom-software or modernization case around a system that already runs: an inventory of the core and the work around it, a named source of truth, fourteen design tests, and a bounded first case that can be stood behind. It will not promise a twelve-week go-live, will not recommend a program to replace the core as a first case, and will not recommend an unsupervised "agentic overlay" that writes to the core as a first case. The honest default is a bounded interface — often read-only first — or a determined path the tools you already own cannot express honestly. Custom software earns its place when "build, buy, or leave as-is" has been asked; the AI Vendor Evaluation Scorecard is the companion for that question. A later increment that replaces or wraps one named component, while the system stays live, with a named cutover and a way back, is a later, different project. This playbook will not pre-clear it, and will not invent a modernization timeline as a company commitment. A person sits on every write that changes a record in the core, and on every customer-facing determination. Review is defined in writing; a prompt is not a gate. Hosting is decided before the case is funded. Where data cannot leave, AsymiLink AI runs on your own hardware, with an optional air gap. Isolation is not coverage of a design that was never finished. We will not claim a certified connector to a named vendor; the core is named as you run it. How the tests and the method are meant to be used This playbook complements the offline-deployment guide, the RAG architecture guide, and the vendor-evaluation scorecard; it does not replace them. The Agentic Workflows white paper is the companion when an overlay is proposed as an agent. The financial method is The Complete Guide to AI Automation ROI: a named process, a baseline agreed before work starts, and a calculation you can defend. This playbook will not invent a modernization-sector return, a "legacy cost" figure, or a twelve-week payback. The 3.8x average first-year ROI across client engagements is an average, not a forecast or a commitment for any individual project. ROI is measured across client engagements against a baseline agreed before work starts. It is an average, not a forecast for any modernization project, and is not used as a sector benchmark. Built to SOC 2 standards is the exact wording we use; it is not a certification of the reader's system, and not a finding that a custom application is safe to run unattended. An illustrative mid-size company with a long-running core, introduced in this playbook and labeled hypothetical, sits at 19, Developing, on the eight-dimension readiness model. Data, Infrastructure and Strategy are relative strengths; Process Automation around the core is a constraint. Its honest first work is a bounded interface or a determined path, with a next-step target of 24, still Developing — a next step, not a timeline, and not a promise. A companion workbook, asymilink-custom-software-playbook.xlsx, holds a family register, the fourteen tests, questions for counsel, a baseline sheet, and the illustrative company. Six families of custom-software and modernization work each with an honest first case, a case that is too wide, and a gate. Each item below is the shape of the map; the full wording, honest first cases and “too wide” warnings are in the PDF. F1. A bounded interface to a system that already runs F2. A determined path the existing tools cannot express F3. A modernization increment of one component, with the system staying live (later) F4. A recommend-and-review overlay (later F5. Unsupervised form refused as a first case) F6. Hosting, decided before the case is funded F7. A portal or customer-facing application, with a person on every determination The PDF gives, for each family, an honest first case, a case that is too wide, and a defined gate. The summaries above are the shape of the map, not a substitute for that wording. Fourteen design tests (excerpt) Fourteen design tests are applied to every case. The answers will differ. The standard does not. The full wording is in the PDF and the companion workbook. T1. including a named source of truth, a write to the core as a different design from a read, a way back on a cutover, a named operator and a stop, and a refusal of a twelve-week rewrite as a first case. A completed test sheet is not a certificate, not a compliance score, and not a finding that a system is safe or aligned with a named regime. Frequently asked questions Q: What is a good first custom-software or modernization case? A: A bounded interface to a core that already runs — often read-only first — or a determined path the existing tools cannot express, with a named operator, a trail, a stop, a person on every write to the core, and a signed baseline. A program to replace the core, or an unsupervised overlay that writes to it, is the wrong first case. Q: Will this playbook promise a twelve-week go-live? A: No. A first case takes the time the fourteen tests demand. This document will not invent a deadline, a "legacy cost" figure, or a twelve-week payback, and it will not treat a named modernization pattern as a timeline we will commit to. Q: Do you claim certified connectors to named cores? A: No. The core is named as you run it. An interface is a design for this case. We will not claim a certified connector to a named vendor. Q: When is custom software the right recommendation? A: When "build, buy, or leave as-is" has been asked, and the path cannot be expressed honestly in the tools you already own. A recommendation can be a tool you already own, a new product, a custom build, or no build at all, and we explain why. Q: What does "Built to SOC 2 standards" mean here? A: It is the exact wording we use for how we build, with role-based access control and AES-256 encryption. It is not a certification of your system, and not a finding that a custom application is safe to run unattended. Q: How is the reported first-year ROI used? A: The 3.8x average first-year ROI across client engagements is an average, measured against a baseline agreed before work starts. It is not a forecast for any individual project, and not a modernization-sector benchmark. This page is the public summary. The full PDF and companion workbook are available through the request form on the page. There is no public file URL that bypasses that form.