Article 02
When Excel should become a system, and when it should not
Assess multiple files, version confusion, unclear access, slow reporting and handoff risk before investing in a system.
Many files
Example situation
Start with the people using the system
Who has the latest file?
The same report exists as Final, Final-2 and Final-revised, while branches use different columns. When management asks for totals, the team spends hours finding files and deciding which numbers are trustworthy. Excel is not the problem; the workflow has outgrown file-based control.
Summary for business and everyday users
When Excel should become a system, and when it should not
Excel still works well for exploration, analysis and clearly owned data.
Risk rises with many people, files, approvals and audit needs.
Define data and ownership before building software.
Interactive explanation
See how the stages connect before reading the detail
Select a stage to understand what happens, what evidence to inspect and how one decision affects the next stage.
Step 01 / 03
01 / Warning signs
The problem starts when a file becomes the workflow
When status, approval, evidence and reports all live in a file, correctness depends on naming, discipline and memory. A system moves those rules into permissions, validation, status and traceable audit history.
Evidence to inspect
- Latest file is unclear
- Many people edit the same data
- Formulas or columns change silently
- Manual consolidation is required
In-depth explanation
Work through each issue in real operating context
Each section connects business impact with what a Tech Lead needs to inspect, including examples, evidence and constraints.
01 / Warning signs
01The problem starts when a file becomes the workflow
When status, approval, evidence and reports all live in a file, correctness depends on naming, discipline and memory. A system moves those rules into permissions, validation, status and traceable audit history.
- Latest file is unclear
- Many people edit the same data
- Formulas or columns change silently
- Manual consolidation is required
- Changes are not traceable
02 / Readiness
02Agree on what the data means before designing screens
Software cannot automatically resolve data definitions that teams disagree on. Define names, sources, owners, validation rules and completion states first. This reduces requirement churn more than starting from polished mockups.
Example
Sales can mean several different numbers
Sales may mean before discount, after discount, after returns or cash received. Without a definition, a new dashboard only displays disagreement faster.
03 / Migrate with control
03Start with a critical flow, not every spreadsheet
Choose a measurable workflow with a clear owner, clean only the required data and define import and parallel-run periods. Expand after users validate the flow. This avoids building a large system on unstable requirements.
Tech Leads should prepare data profiling, mapping, duplicate rules, migration rehearsal, reconciliation reports and rollback plans.
Checklist before action
Name a data owner
Agree on core definitions
Identify the source of truth
Prepare sample data
Choose a clear first flow
Plan import and reconciliation
References
SIS working method and calculation notes.
Figures and examples create a discussion framework. Validate them against the real system and its constraints before deciding.
- Reviewed by
- SIS Product & Engineering
- Last reviewed
- 2026-07-17
Not sure where to start the review?
Use a preliminary tool or share the system context with SIS so the highest-priority work can be identified.