EAFLOW · SOLUTIONS · CROSS-INDUSTRY SOLUTION

Departmental Application Modernization

A spreadsheet, a legacy departmental database or a prototype built with AI already holds real rules and criteria validated in day-to-day use. That work is recovered and turned into a process with forms, tasks, owners, integrations and an operating model the organization can sustain.

Category
Departmental Application Modernization — cross-industry solution on the Operational Graph
When to use
A process that today lives in a spreadsheet, a departmental database, an internal procedure or a prototype, and that now needs named owners, integrations and an operating model the organization can sustain
Users
The business area that owns the process, the IT owner who cannot support that application today, and everyone involved in the flow: requesters, reviewers, authorizers and operations
Output
A working process with forms, tasks, explicit rules, integrations and exception handling, plus an agreed model for operation, maintenance and evolution with named owners
Implementation
Cross-industry solution with assisted implementation; it can run alongside the current application through the agreed transition
Does not replace
Existing rules are not migrated automatically: each one is reviewed with the team that uses it and agreed before implementation. This does not replace the ERP or the systems of record: the process stays connected to them, and closure is confirmed either by the authorized source or by the authorized owner.

01 · The problem

The application still works; the operating model around it is no longer sustainable

A team built what it needed: a spreadsheet holding the approval rules, a local database with years of requests, an internal procedure, or a prototype assembled to test an idea. Those pieces solve a real problem, and they hold criteria nobody else ever wrote down.

The approval rules live in Microsoft Excel formulas that only one person can read; the history sits in a Microsoft Access database with no documented backup; the procedure is spread across Microsoft SharePoint lists and email threads, with a separate copy of the file for each person. The Microsoft Power BI dashboard shows the backlog: it reports status, but it does not assign the work or leave a record of who approved what.

The business area can no longer change a rule without asking someone for a favor. IT takes support calls on something it never designed, with no documentation, no test environment and no way to answer for its availability.

The starting point is not a mistake. It is a ceiling: what stopped scaling is not the rules, it is the operation built around them.

02 · The solution

From the rules that already exist to a process the organization can run

The work starts by recovering what already works: which fields users complete and how each is validated, which conditions trigger an approval, and which exceptions are still handled by hand. That content is worked through with the team that uses it and checked against the historical cases held in the application. No rule is migrated automatically: a rule nobody can explain is recorded as an open decision rather than copied blindly.

The flow is then redesigned on top of those rules: forms with the agreed validations, tasks with an owner and a deadline, authorizations by rule and by amount threshold, and statuses that separate what was requested from what was executed. BPMflow runs the workflow — cases, assignment, deadlines, authorizations and evidence. The Operational Graph supplies the surrounding context — legal entity, business area, responsible role and the documents that apply — so the rules rest on current definitions approved by the organization rather than on a copy buried inside a form.

What counts as closure

  • A recovered rule is not yet an agreed rule. The spreadsheet shows how the work is done today; the business owner decides whether that is still the rule. What gets implemented is what was agreed, with a named owner.
  • A form that replaces the spreadsheet does not yet replace the application. Replacement is defined by scope: what moves to the new process, what stays where it is, what remains out of scope, and how long the old path keeps running alongside it.
  • A task marked done is not a confirmed outcome. Each flow agrees who or what confirms it: the record returned by the responsible system, or the authorized owner where the outcome does not live in a system.

When what needs modernizing is a repository of process or architecture models — diagrams, legacy notations, attributes and relationships — the route is a different one: Legacy Modernization and Process Knowledge. When the process starts from a deviation that today shows up only on a dashboard, the way in is From Reporting to Action.

03 · Starting points

Where a departmental modernization starts — spreadsheet, legacy database or prototype — and what changes in each

The starting point changes how much can be reused on day one, not where the work ends up. And it does not always start from something already built: the same approach applies to creating a capability that does not exist yet, or to replacing a legacy application whose scope is already well understood.

Spreadsheet and email

The calculations and the criteria live in the spreadsheet; everything else happens over email.

What is kept

The criteria, the calculations and the categories the team has already validated in day-to-day use.

What changes

A single instance of each case, with a named requester, reviewer and authorizer, and no parallel copies to reconcile.

Legacy departmental database or application

Forms, logic and years of data, with permissions that accumulated one exception at a time.

What is kept

The historical data and the logic that still holds; the application can keep running through the transition.

What changes

What is kept, what is adapted and what is replaced is decided as part of the scope, with permissions reviewed and every integration documented.

Internal application with no maintenance owner

It works, but nobody today can change it or answer for its availability.

What is kept

Its real scope: what it resolves, who depends on it, and what happens if it becomes unavailable.

What changes

An owner is assigned for operations and ongoing change. The scope is then either sustained or replaced by an explicit decision, never by neglect.

Prototype built with AI

It already demonstrates the idea; what is missing are the conditions to run it day to day.

What is kept

How the flow was designed, the product decisions behind it, and what the team learned while testing it.

What changes

Integrations, permissions, error handling and testing are completed, and a support model is put in place. The prototype is not thrown away.

04 · Capabilities

What is in place once the spreadsheet stops being the system

The business area gets back the ability to change a rule through an agreed procedure, and IT takes on an application that comes with documentation, permissions and named owners. Max answers questions about a case within the authorized context: it shows the status, the evidence on record and the rule that was applied.

  • Recovering the rules and criteria

    Fields, validations, approval conditions and exceptions are worked through with the team that uses them, and checked against the historical cases already held in the application.

  • Forms and tasks for the redesigned process

    Forms with the agreed validations, tasks with an owner and a deadline, and statuses that distinguish what was requested, what was authorized and what was confirmed.

  • Integration with the systems of record

    The process pulls in the context it needs and writes the outcome back where it has to be recorded, with the confirmation agreed for each step.

  • Exception and error handling

    If a system does not respond or a value fails validation, the case records and keeps an open issue visible, with an owner and a route to retry or resolve it manually.

  • Coexistence and an agreed transition

    The new scope can run alongside the current application, with an agreed cut-over criterion, testing on cases from the current period and a rollback plan.

  • Operation, maintenance and evolution

    The agreement records who operates the process, who changes a rule, how that change is approved and what happens to the cases already in progress.

05 · Roles

The rule, the authorization, the support and the daily work each have a different owner

Business area owner

Provides the rules and decides which ones still apply.

What they see

The process with its rules written out, the cases in progress and the open items in their area.

What they do

Explains how the work is done today, agrees which rule gets implemented, and requests later changes without depending on one individual.

Author of the spreadsheet or application

Knows the detail that was never written down.

What they see

The recovered rules, and the cases where current behavior does not match what was written down.

What they do

Corrects the interpretation, supplies the history, and validates the flow against past cases.

IT owner

Takes on what IT cannot support today.

What they see

The agreed integrations, the permissions, the environments and the open technical items in scope.

What they do

Agrees on access and approved source systems, participates in testing, and accepts the process handover with its documentation and support model.

Process user

Does the day-to-day work.

What they see

Their own tasks, the committed deadline and what information each case is still missing.

What they do

Completes the form, addresses review comments and provides the confirmations assigned to them.

06 · Implementation

The work starts from the application in use and its recent cases

The application in use today, its recent cases and the team that operates it are enough to start: no prior architecture project and no complete process documentation are required.

1 · Agree scope and outcome

Which process is in scope, and what counts as complete

Which rules stay exactly as they are, which ones need to be revisited and reapproved, who authorizes, and what counts as a confirmed outcome.

2 · Recover the rules and connect the sources

Fields, validations, data and integrations

Fields, validations and exceptions are drawn out of the current application and its historical cases; the context sources and the system that records the outcome are agreed.

3 · Test, run in parallel and hand over

Testing, coexistence and operation

The process is tested with cases from the current period, rules and deadlines are tuned, the cut-over criterion is agreed, and the client team takes over day-to-day operation.

Illustrative example

Investment requests in a local database, approved over email

The database holds the form, the numbering, the amount calculation and an approval rule by amount threshold that only its author fully knows; the authorization sits in email. The redesigned flow keeps the form and the thresholds, records them as rules of the process, and connects them to the legal entity, the cost center and the owner currently responsible.

The exception that matters appears when the amount changes after approval. The case does not start over. If the revised amount crosses a threshold, the case is routed to the authorizer for the new approval band, while the earlier approval and its rationale remain on record. Through the transition the database stays available read-only until the open cases finish on the old route. Closure is confirmed by the record returned from the system where the investment has to be booked.

The example describes how a scope is framed. It does not describe an existing deployment.

07 · Typical scenarios

When modernizing is worth it, even at low volume

A low-volume process still justifies the work when the cost of an error is high, or when the operation depends on a single person.

  • Spreadsheet with approval rules by amount threshold
  • Departmental database with years of history
  • Procedure split across lists and email
  • Internal application with no maintenance owner
  • Prototype that needs integrations and permissions
  • Process with no record of who approved it
  • Rules only one person can explain
  • A new capability no system covers yet

A departmental application is modernized when its rules are explicit, the process has named owners, and the outcome is confirmed where it belongs. The practical way in is one specific process, starting with the rule only one person can explain.

Departmental Application Modernization is an EAFlow cross-industry solution on the Operational Graph, supported by BPMflow automation capabilities. Legacy Modernization and Process Knowledge cover model repositories; From Reporting to Action covers a deviation spotted on a dashboard. The examples on this page are illustrative.