The ERP transformation toolkit
A practical library for the whole ERP lifecycle, from deciding whether to replace a system through selection, delivery, and the first weeks after go-live. Filter by where you are in the journey, what kind of resource you need, your function, or your industry.
52 resources
Scores business readiness, process standardisation, data quality, executive alignment, internal capacity, and technical complexity, then returns a readiness grade with the gaps to close before a program starts.
What to put in place in the twelve months before an ERP program starts, so the program is not the first time these questions get asked.
Clean core means keeping custom code out of the ERP so upgrades stay cheap. Composable means buying best-fit pieces around it. What you are really buying, and the discipline it demands.
Running Dynamics 365 or NetSuite in the subsidiaries under an SAP or Oracle core can cut subsidiary cost and time, if the consolidation interface and chart of accounts are governed centrally.
The work that makes a program run clean happens before the integrator arrives: data cleanup started, process owners freed, the business case owned, and the start date set by the 36-month rule.
A weighted scoring model that turns demo impressions and requirement fit into a defensible ranking, with the weights, scores, and the gap between finalists all shown.
A structured Excel catalog of common ERP requirements across finance, procurement, inventory, order management, and manufacturing, pre-populated so you refine rather than start from an empty sheet.
The downloadable version of the scorecard: a weighted Excel model with criteria, per-evaluator scoring, demo evaluation, and an automatic weighted ranking across up to five vendors.
What belongs in an ERP RFP and what to leave out: the sections that get useful answers, the scenarios that expose a fit gap, and the commercial questions vendors answer differently when they are in writing.
The full selection sequence: requirements, longlist, shortlist, demos scored on the same scenarios, references, and a recommendation the board can follow.
A ranked shortlist from twelve vendors based on your manufacturing mode, plant count, revenue, and whether an MES is already in place.
A ranked shortlist weighted on asset management, work management, meter-to-cash, and regulatory reporting, matched to your segment and rate case calendar.
No utility ERP vendor leads all four scoring areas. How your size and rate case calendar pick the shortlist, and why only SAP and Oracle own a billing engine.
A defensible software choice comes from fit, not a vendor 'alternatives' list: the right tier for your size, named conditions for each vendor, and reference customers of your shape.
Below about $500M and ten plants, a tier 2 system covers the operation for roughly half the cost and in a third of the time. When SAP or Oracle is overkill, and when it earns its price.
Six questions decide a utility ERP program before the RFP: CIS sequencing, the FERC chart of accounts, rate-case timing, asset-data cleanup, the field-service test, and the segment shortlist.
At $500M to $5B, both tiers are live and the budget differs 3x to 10x. The conditions that decide it are entities, countries, volume, and five-year growth, not revenue.
Shop-floor and MES integration is the largest cost driver in a mid-market manufacturing program, ahead of finance. The MES decision belongs in the vendor choice, not after it.
How to choose the firm that implements the software: committed people, an estimate that covers the whole job, methodology that fits, and references of your shape and size.
What to ask an integrator in writing and how to read the answers: named staffing and tenure, what the fixed price does and does not include, and the client responsibilities hiding in the assumptions.
The software and implementation terms that protect you: deliverable acceptance, key-person tenure, change control, and the leverage you lose the moment selection narrows to one.
The integrator is 40 to 60% of the bill and the biggest driver of whether the program lands. Choose on the named team, references of your shape, rate transparency, and the contract that holds them.
AI removes junior hours first, so hours fall faster than the bill and the blended rate rises. Even a real 50% effort cut is worth 20 to 30% of budget only if the contract makes the SI pass it through.
Three contracts decide your downside: the software, the integrator, and the cloud. The clauses that protect you, and why you time the signature to the vendor's quarter-end, not the support date.
Ten questions expose padding in any SI proposal: the hours baseline, the blended rate, the assumptions, and the change-order triggers. The SI is 40 to 60% of the bill.
A low, mid, and high program range for your shortlisted vendor tier, rebuilt from the published ratios with every input and multiple shown.
A working Excel model that totals software, implementation, integrations, data, dual-run, internal labour, contingency, and five-year run cost, with the ratios built into the formulas.
Every line a defensible ERP budget needs, from software and implementation to the items that blow up late: dual-run, data cleanup, integrations, and run cost.
How to build a case the board will approve and still recognise in year three: the benefits with owners, the costs with ranges, the assumptions register, and the decision the case is really asking for.
Budget 1x to 3x first-year software to implement, and 3x to 5x over five years. The ranges by tier and company size, with the arithmetic to rebuild any number an integrator gives you.
Six line items cause most ERP overruns, whatever the vendor: dual-run, data cleanup, integration, custom code, testing, and change. What each costs and how to cap it.
Build the budget from the ratios, not the vendor quote: 1x to 3x software to implement, 15 to 25% a year to run it, plus the six line items that overrun. Then present the five-year total.
A 12-month delay rarely saves money. It adds support premiums, compresses the program, and pushes past the vendor incentives that cut price, often more than $1M on a mid-size program.
The board judges the program against the number you gave it. Present the five-year total, not the go-live figure: software, implementation, run cost, and dual-run, with every assumption shown.
The roles an ERP program needs on the client side, the ones you must not outsource to the integrator, and the signs a role is missing.
Who decides what, and who is merely consulted: a governance model and a responsibility matrix that fixes decision rights before the first disagreement, not after it.
A structured way to map who is affected, who has influence, and who will resist, so the communication and change effort goes where it changes the outcome.
The phases of an ERP program, what each produces, how long each takes and what moves it, so you can build a plan that holds instead of accepting the integrator’s dates on trust.
A working Excel planner for the work that is always underestimated: a data object inventory with owners, cleanup tracking, a load sequence, and a reconciliation check that totals source against target.
AI is cutting work packages, not programs. What is real, what is a demo, and the five questions that make an integrator put the saving in the contract rather than the sales deck.
An agent-drafted configuration your auditors cannot trace is a control gap. The approval gates, audit trail, data-access limits, and ownership terms to require on any platform before agents touch configuration.
The roles you must own and not hand to the integrator: the sponsor, a client program manager, business process owners, a data lead, and independent assurance. The gaps that sink programs.
A 20-question self-assessment that scores your program against the seven signs and returns red, amber, green by workstream, with an AI-readiness section.
The test of whether the business, not just the system, is ready: can you still receive, ship, invoice, pay suppliers, and close the books the week after cutover.
Seven measurable signs lead the status report by about two quarters. The thresholds, the records to ask for, and when three of them mean you should commission an independent review now.
Budget 1 to 2% of the program for independent assurance that reports to you, not the program. It reads the records the status deck summarizes, and it pays for itself by catching drift two quarters early.
A new SI inherits the same governance, data, and staffing problems plus a learning curve. A rescue costs 30 to 50% of the original budget, and most of the fix is governance, not a new contract.
A five-year cost comparison of migrating now, extending support, or switching vendors, built from your own maintenance base and your vendor’s actual dates.
Only one major ERP support deadline is both close and firm. What the SAP, Oracle, and Microsoft dates really mean, and the date 36 months before each one that actually decides your timeline.
SAP's 2027 date is one deadline with three decision points, 2027, 2030, and 2033. The three real options, what each costs, and when you must choose.
Oracle has committed Premier Support for EBS 12.2, JD Edwards, and PeopleSoft through at least 2037, so the move is a business case with no external clock.
Dynamics AX is already unsupported and Dynamics GP support ends in 2029. What running past the date actually costs, in security, compliance, and a compressed replacement.
No resources match these filters. to see everything.