Software & ERP implementation methodology
The delivery discipline behind every Watkyn engagement
Modernization projects fail in predictable ways: vague requirements, underestimated data migration, communication that disappears mid-build, weak adoption, and a system that ends up depending on one overextended person. Watkyn’s six-phase delivery methodology exists to prevent those failures. It gives government agencies, higher education institutions, nonprofits, and growing businesses a documented path from discovery through support, with written sign-offs, staged migration, continuous testing, training, and steady communication built in.
We apply the same discipline across everything we deliver: custom software development, Acumatica ERP implementation, Quickbase and Caspio development, Amplify GenAI implementation, and AI governance work, and we right-size it for app rescue and spreadsheet replacement.
1. Discovery → 2. Design → 3. Build → 4. Test → 5. Deploy → 6. Support
Proven on: Colorado OADC system consolidation · CCR REN regional energy CRM · City of Philadelphia economic-development software · Saint Louis University platform built for an 18,000-student campus
Built for government, higher education, nonprofits, and accountable enterprises
The organizations we work with differ in size and sector. What they share is accountability. Each one answers to someone for how its systems run, whether a procurement office, a board, a funder, an auditor, a regulator, or customers who can’t tolerate downtime. A methodology built to satisfy government scrutiny serves all of them.
Government agencies
State and local agencies and special districts operating under formal procurement, public scrutiny, and structured documentation requirements.
Colleges and universities
Institutions modernizing financial and operational systems, governing AI adoption, and protecting student data under FERPA and accreditation pressure.
Nonprofit organizations and associations
Mission-driven organizations that need dependable systems for reporting, approvals, program operations, finance workflows, and board oversight, without piling on administrative work a lean team can’t sustain. For nonprofit ERP work, our methodology places special emphasis on fund accounting, grant tracking, restricted funds, and reporting obligations.
Mid-market organizations and enterprises
Companies replacing end-of-life ERP or outgrowing the systems they started on, who want a partner that documents the work and stays accountable for it.
Running on a Quickbase app that stalled when its builder left, or operations that have outgrown a spreadsheet? The same six phases apply, right-sized. See Quickbase App Rescue, Outgrowing Spreadsheets, or the Small & Midsize Business industry page.
What disciplined delivery prevents
Evaluating a modernization partner is really an exercise in evaluating risk. Here is the risk each part of our process is built to remove.
Where projects go wrong
How our process controls it
Vague requirements
Requirements are documented and signed off before any build begins, so everyone has agreed in writing on what the system must do.
Data surprises
Data is inventoried in discovery, then staged, reconciled, and validated against the source system before anyone relies on it.
Black-box development
You see working software at a steady cadence, and a short written status update covers progress, risks, and what comes next.
Failed adoption
Training and full documentation are part of deployment, so the people who run the system are ready on day one.
Single-person dependency
Decisions, specifications, and runbooks live in a durable, organized project record that future staff or a future vendor can pick up and read.
Security and access-control gaps
Permissions, roles, data protection, and access control are designed into the architecture, and changes that affect security or access are reviewed before they reach production.
Security runs through every phase. See Security & Compliance for our practices.
Proven where the work has to withstand scrutiny
This methodology wasn’t designed in the abstract. It was developed and refined on engagements where documentation, accountability, and continuity were contract requirements, and it has held up over client relationships measured in years.
Colorado Office of the Alternate Defense Counsel: Engaged to consolidate four legacy government applications into a single cloud platform on AWS, with a large-scale document migration, role-based access control, and delivery under a formal government contract with structured documentation and change control.
Central California Rural Regional Energy Network (CCR REN) / County of San Luis Obispo: Built a custom CRM for a regional energy network on AWS serverless architecture, including compliance-metrics reporting, customer engagement, and outreach campaign management.
City of Philadelphia, Department of Commerce: A department-wide economic-development and relief platform, supported continuously from 2020 through 2026, a six-year engagement.
This work was performed for The City of Philadelphia and its Office of the Director of Commerce.
Saint Louis University: Health-compliance and operations systems for an 18,000-student university, delivered and expanded over a 42-month engagement; the pandemic-era platform was kept and made permanent student-health infrastructure.
Nonprofit Organization: An ongoing engineering, data, and AI partnership since 2023 with a leading national civil rights nonprofit, supporting workforce development, housing counseling, and education programs across a national affiliate network, a relationship that has grown every year.
Group Travel Odyssey (Kaleidoscope Adventures): A travel-operations ecosystem built and supported since 2018, Watkyn’s longest-running client relationship, spanning role-specific portals over a shared data layer, payment processing, secure document upload, and two-factor authentication.
For full government past-performance detail, including scope, duration, and environment, see our Government page. For institutional work in depth, see Higher Education; for mission-driven and association work, see Nonprofit & Association; for project narratives, see Custom Software Development.
What you have at the end, and all along the way
Documentation isn’t produced at the finish line; it accumulates as the work proceeds, which is what lets a new staff member, an auditor, or a future partner understand the system without reverse-engineering it. A typical engagement leaves you holding:
Requirements specification · project plan and timeline · system architecture · data model and data dictionary · data migration plan · integration plan · security and access-control design · test plan and UAT record · training materials · user and administrator documentation · go-live checklist · post-launch support plan.
Throughout the project you also receive change logs and status reports at set intervals, the running record public contracts and institutional accountability require.
A project record that outlives the project
On engagements delivered under formal contracts, everything the project produces lives in a version-controlled repository alongside the source code: numbered architecture decision records, specifications, tickets, meeting summaries, onboarding guides, playbooks, and runbooks. Documentation is treated the way code is treated, with history, authorship, and review.
The practical benefit is durability. The whole project can be replayed: why each choice was made, what shipped when, and how the system operates. You inherit working software and the full context behind it, a searchable knowledge base that outlives any individual and removes vendor knowledge lock-in.
You always know where the project stands.
Every engagement opens with a kickoff and a defined agenda. From there, we demonstrate working software on the cadence you prefer, weekly or biweekly, and send a short written status update on the same rhythm: progress, risks, and what comes next. When an issue needs executive attention, there’s a defined escalation path. Any change to scope, schedule, or cost goes through a formal change-request process, and a shared risk-and-decision log stays current throughout. We track technical work in a dedicated project tracker, and the scheduled status report keeps progress visible without anyone having to chase it.
The cadence:
kickoff with agreed agenda
working demonstrations and written status updates on your preferred rhythm
recurring meetings with your project lead
a defined escalation path
change control with a shared risk-and-decision log
Iterative progress, governed delivery
You’ve probably heard “agile,” “sprints,” and “waterfall” before. We work in a lean, continuous flow with formal phase gates: working software is demonstrated at a steady cadence, a scheduled written status report keeps progress visible, and every phase closes with a documented decision or acceptance point; you sign off wherever scope, design, operational risk, deployment, or acceptance is yours to approve. That combination gives you the predictability of a phased plan and the visibility of frequent working demonstrations, without a heavy meeting calendar.
We test throughout instead of at the end and produce working software in every iteration, so value accumulates steadily rather than arriving in one large integration at the finish; on custom software builds, the build-and-deploy pipeline goes up in the first weeks. Where a contract calls for milestones, we prefer to tie them to demonstrable, working software. Iterative build with procurement-grade documentation is what makes the work both quick to see and safe to audit.
Why our size works in your favor
Watkyn is intentionally compact. Senior people stay close enough to understand the details, while the process is documented enough that you’re never dependent on one person’s memory. That gives institutions and nonprofits a useful middle ground: more continuity and structure than a freelancer, with more direct accountability than a large consultancy. We’ve operated this way for many years.
Same six phases, right-sized to the work
The framework is one discipline applied at different depths. Here’s how it shifts.
By engagement type
Custom software development
Phase 2 (Design & Architecture) includes full-stack architecture design; Phase 3 (Development & Integration) is application development built from the ground up, under the documentation and change-control standards public contracts require. We build cloud-native systems on AWS and also deliver in client-mandated environments, including Microsoft .NET and SQL Server stacks; when a procurement specifies the platform, we design to it and confirm compliance requirement by requirement. See Custom Software Development.
Acumatica ERP implementation
ERP work puts extra weight on the parts that most often sink projects: chart-of-accounts and fiscal-calendar setup, approval workflows, role-based permissions, reporting requirements, third-party integrations, data cleansing, and parallel-run reconciliation before cutover. Phase 2 (Design & Architecture) expands to cover financial configuration; Phase 4 (Testing & Validation) validates against real financial data. See Acumatica ERP Implementation for ERP-specific migration and readiness detail.
Quickbase development and app rescue
Quickbase work runs all six phases in a compressed, fixed-price form. Discovery and design still happen, sized to the job, so even a small build gets a sound data model, real permissions, and documentation. App rescue starts with a fixed-fee assessment of the app you already have. See Quickbase Development and Quickbase App Rescue.
Caspio development
Caspio projects follow the same compressed, fixed-price form, with extra weight on portal design, role-based access for external users, and integrations with the systems your operation already runs on. See Caspio Development.
Amplify GenAI implementation
Discovery identifies your use cases, data sources, and governance obligations. Design covers the deployment architecture in your own AWS account, identity and single sign-on, and usage and cost controls. Build, test, and deploy stand the platform up on your approved sources, and support hands it to your internal team with documentation and training. See Amplify GenAI Implementation.
AI governance
Discovery identifies current AI use, risk exposure, policy gaps, and the people who need to govern it. Design shapes the governance structure: policies, review processes, and alignment with frameworks such as ISO 42001 and the NIST AI RMF. Implementation trains the people involved and brings the process into day-to-day operation. For governed AI infrastructure in your own environment, see Amplify GenAI Implementation.
By organization type
The phases don’t change by sector, but the emphasis does. Government work leans hardest on documentation, change control, and procurement-grade deliverables; see Government for public-sector past performance. Higher education adds data protection and committee-paced review; see Higher Education. Nonprofits and associations put the weight on fund accounting, grant and restricted-fund tracking, program and donor reporting, and board oversight: dependable systems that don’t add administrative burden a lean team can’t carry. See Nonprofit & Association. Mid-market and enterprise work moves faster and trades some formality for speed, while keeping the documentation that prevents single-point dependency. Small and midsize businesses get the compressed, fixed-price form described above; see the Small & Midsize Business industry page.
Common questions about how we work
Every engagement starts with Discovery & Requirements: we interview the people who depend on the system, audit the current environment, inventory the data, and produce a written requirements set you sign off on before any build begins.
A deliberate hybrid. We work in a lean, continuous flow with formal phase gates: working software is demonstrated on your preferred cadence, written status reports keep progress visible, and each phase closes with a documented decision or acceptance point, with your sign-off wherever the decision is yours to make. You get the predictability of a phased plan and the visibility of frequent demonstrations, without a heavy meeting calendar.
Requirements are documented and signed off in Phase 1 (Discovery & Requirements), and each phase has a clear decision point. Changes go through a formal change-request process and are logged in a shared risk-and-decision record, so scope is managed deliberately and every change is a recorded decision.
Data is inventoried and assessed in Discovery, staged and prepared during development, then validated through a staged migration and parallel-run verification, confirming the new system matches the source on real data before anyone relies on it.
A typical engagement leaves you with a requirements specification, system architecture, data model and dictionary, migration and integration plans, security design, test and UAT records, training materials, user and administrator guides, a go-live checklist, and a support plan, plus change logs and status reports throughout.
A kickoff with a defined agenda, working demonstrations and a short written status update on your preferred cadence so you can see progress without asking, recurring meetings with your project lead, and a defined escalation path.
Yes, openly. Our senior developers use AI to plan builds, think through data models and architecture, and accelerate delivery. The judgment and the accountability stay human: an expert designs, reviews, and stands behind every system we deliver. We do not use client data to train AI models, and we use AI tools and settings designed to prevent submitted client data from being used for model training. It is the same discipline we apply in our AI governance practice for institutions that must answer for how their systems behave.
Usually, yes. Our cloud builds run on AWS, and we also deliver in client-mandated environments, including the Microsoft stack: C#, .NET, SQL Server, and IIS. When a procurement specifies the platform, we design to it and confirm compliance requirement by requirement.
Yes. The same six phases run in a compressed, fixed-price form. Discovery and design still happen, sized to the job, so even a small build gets a sound data model, real permissions, and documentation.
Yes. We work with mission-driven organizations and associations that need dependable systems for reporting, approvals, program operations, finance workflows, and board oversight. For nonprofit ERP work, our methodology places special emphasis on fund accounting, grant tracking, restricted funds, and reporting obligations.
The six phases stay the same, with extra weight on financial configuration: chart of accounts, fiscal calendars, approvals, permissions, reporting, integrations, data cleansing, and parallel-run reconciliation before cutover. See Acumatica ERP Implementation.
Planning a system modernization?
Start with a short call. We’ll identify your current system, the risk points, and the likely sequence. If there’s a fit, we can turn that into a structured readiness review before anyone talks scope or budget.
Questions, or something specific you’re weighing? We’re happy to talk: 201-644-5251.
Working in a regulated or high-scrutiny environment? See Security & Compliance.
