Skip to content
Tutorials Center
My learning

Topics

Lesson 19 of 36 ยท Implementation and Change

The ERP and CRM Implementation Lifecycle

Walk through the phases of an ERP or CRM implementation, from discovery to hypercare, with the deliverables, gates and risks of each phase and a small plan checker in Python.

Buying business software is the easy part. Making it work inside a real company, with real people, messy data and old habits, is the hard part. Studies of enterprise projects keep finding the same thing: most overruns and failures come from planning, people and data, not from the software itself. A clear implementation lifecycle gives everyone the same map of what happens, in what order, and what "done" means at each step.

This lesson describes a typical lifecycle that applies to ERP and CRM projects alike. Vendors give the phases different names (SAP Activate, Oracle's AIM successors, Microsoft's Success by Design, Salesforce's own methods), but the ideas underneath are the same. The product is not available on this site, so the examples here use small Python stand-ins to model plans, dependencies and gates.

The phases at a glance

Before the build

Discover Goals, scope, current processes.
Plan Team, budget, schedule, risks.
Design Future processes and fit-gap.

Cheap to change now, costly to change later.

1 / 3

1. Discovery and project preparation

Discovery answers "why are we doing this?" Typical outputs are a business case (expected benefits and costs), a scope statement (which companies, sites, modules and processes are in and out), a list of stakeholders, and the success measures, for example "close the month in 5 days instead of 12" or "raise lead-to-opportunity conversion by 20 percent". Projects that skip measurable goals cannot prove they worked.

Preparation also sets up governance: a steering committee of executives who decide on money and scope, a project manager, process owners from each department, and technical leads. Without named owners, decisions stall.

2. Planning

The plan turns scope into work: a work breakdown structure (WBS), a schedule with dependencies, a budget, a communication plan and a risk register. A good risk register lists each risk with a probability, an impact and a named person who acts on it.

3. Design (blueprint)

Teams describe the future processes ("to-be"), compare them with what the standard product does (the fit-gap analysis), and decide how each gap is closed: change the process, configure, extend, or build something custom. The output is a signed-off blueprint or solution design. Changes after sign-off go through a change-control board.

4. Build: configuration, extension and data

Consultants configure organization structures (companies, plants, sales areas), master data templates, workflows, roles and reports. Developers build the extensions and the interfaces described in integration patterns. In parallel, a data team extracts, cleans and maps legacy data (data migration).

5. Test

Testing happens in layers: unit tests of single configurations, integration tests of end-to-end flows such as order to cash, performance tests with realistic volumes, security and role tests, and user acceptance testing (UAT) by business users. See testing, cutover and go-live.

6. Train and prepare the organization

Training and communication run alongside the build, not after it. The people side is covered in change management and training.

7. Go-live, hypercare and optimization

After the cutover, the team gives intensive support (hypercare, often 2 to 8 weeks), tracks incidents and fixes defects. Later, the work shifts to continuous improvement: new reports, extra modules, user feedback.

Gates: deciding when a phase is finished

A phase gate is a checkpoint where the steering committee decides to proceed, fix something or stop. Each gate has entry and exit criteria. For example, the exit criteria for the design phase might be: blueprint signed by all process owners, every gap has a decision, and the risk register is up to date.

Python A phase-gate checker (stand-in)

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The checker is deliberately simple: a gate is only useful if the criteria are written down in advance and are checkable. "The system feels ready" is not a criterion.

Scheduling with dependencies

Phases overlap, but some tasks cannot start before others finish. The length of the longest dependent chain is the critical path: any delay on it delays the whole project. The next stand-in computes the earliest finish of each task from a small task list.

Python Critical path of a small plan

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

If "Data migration" grows from 9 to 12 weeks, the critical path stays the same chain and the end date moves by 3 weeks. If "Training" grows by one week, nothing changes, because it has slack. Try both edits and watch the result.

Common risks across phases

  • Scope creep: every department adds "just one more" requirement. Use a change-control board and a backlog for phase 2.
  • Weak sponsorship: executives who disappear after the kickoff. Decisions then wait for weeks.
  • Underestimated data work: dirty master data is the most frequent surprise.
  • Too much customization: every custom object adds cost for upgrades and testing.
  • Late user involvement: users first see the system at UAT and reject it.
  • Compressed testing: when the schedule slips, testing time is cut first. Resist this.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Which output normally closes the design (blueprint) phase?

    The blueprint documents the future processes and how each fit-gap item is resolved. It is signed off before configuration begins.
  2. 2. What does the critical path tell a project manager?

    Tasks on the critical path have no slack. Tasks off the path can slip a little without changing the end date.
  3. 3. Why should success measures be defined during discovery?

    Measurable goals such as faster month-end close let the team judge the result and make trade-offs during the project.
  4. 4. What is hypercare?

    During hypercare the project team stays close to users, fixes defects quickly and monitors the new system.

Summary

  • A lifecycle gives the project a shared map: discover, plan, design, build, test, train, cut over, hypercare, improve.
  • Each phase has outputs and a gate with written criteria.
  • Design decisions are cheap to change early and very expensive late.
  • The critical path shows where schedule risk lives.
  • Data, people and scope cause more trouble than software.

Next: Implementation Approaches: Big Bang, Phased and Parallel