Getting an ERP Selection Approved - The Business Case Your CFO Will Sign
Updated: 14 hours ago
The person who starts an ERP search is rarely the person who signs for it. The search usually begins in IT, procurement or a program office; the signature comes from the CFO, or from the CEO in a smaller company. Between the two sits a business case, and most business cases fail for the same reason: they ask the CFO to approve a large, vague investment on the strength of a vendor's return-on-investment slide.
This guide is for the person writing that case at a manufacturer or distributor. It covers why CFOs say "not yet," how to structure the request so it is easy to approve, what the document should contain, and the numbers that carry weight in a finance review. The short version: ask for a smaller first decision, and show what the business is losing today.
Why CFOs say "not yet"
The benefits are adjectives - "Modern," "integrated" and "real-time" are not numbers anyone can be held to.
The ROI came from a vendor - A vendor's calculator is built to produce a yes. Finance knows it and discounts the whole document.
The cost is only the software - Licenses are often the smaller part of the bill. Implementation, internal time, data cleanup, integration, training and several years of running costs are what a CFO worries about.
There is no baseline - Without today's close days, inventory position or error rates, nobody will be able to say later whether the project worked.
It asks for everything at once - One approval for the selection, the software and the implementation, before anyone knows which system or partner, reads as a request for a blank check.
Ask for two decisions, not one
The most effective change is structural. Split the request into two gates.
Fund the selection first - Approve a defined program of twelve to twenty weeks that produces requirements signed off by the business, a shortlist, scripted demonstrations, a total cost of ownership comparison and a recommended system and implementation partner. The CFO is approving a known scope, a known timeline and a limited commitment.
Approve the investment on evidence - At the end of the selection, the CFO decides on real proposals, a like-for-like cost comparison and a named partner, with the option to stop or to upgrade what the company already has.
A CFO who would hesitate to approve the whole program on the strength of a slide will usually approve a defined selection that ends with better information and a clear exit. Here is how a selection program runs, week by week.
What the business case should contain
The decision requested - One sentence at the top: approve a selection program, with its scope, timeline and the people it needs.
Why now - The events forcing the timing, such as a support end date, an acquisition, a new plant or branch, an audit finding, or a growth plan the current system cannot carry.
What the current system is costing - The baseline, in the measures below, with the source of each number.
Options considered - Including doing nothing and upgrading in place. A case with no alternatives reads as a foregone conclusion.
What the selection will produce, and when - Requirements, shortlist, demonstrations, cost comparison and recommendation, each with a date.
What it needs from the business - Named process owners and their hours. Internal time is a real cost, and the CFO will ask about it.
Risks in both directions - The risk of acting, such as disruption and overruns, and the risk of waiting, such as an unsupported version, key people leaving or growth the system cannot absorb.
Governance - The executive sponsor, the steering committee and the date of the second decision.
The numbers a CFO takes seriously
Choose measures finance already tracks and that a new system can plausibly move. For each one, record today's value, where the number comes from and who owns the improvement.
Month-end close - Days to close, and the manual journal entries and spreadsheet reconciliations behind them.
Inventory - Days on hand or turns, obsolete and slow-moving stock, and cycle-count accuracy, by plant or branch.
Margin leakage - For distributors, unclaimed rebates, pricing overrides and freight that is never billed. For manufacturers, standard costs that no longer match actuals, scrap and rework.
Service - On-time, in-full delivery and backorders, which show up as lost orders long before they show up in the P&L.
Growth capacity - The people the business would have to add at the next plant, branch or acquisition if processes stay manual. This is usually more credible than promising headcount reductions.
Risk and control - Audit findings, segregation-of-duties gaps, an unsupported version, and how many people understand the current system.
Keep the estimates conservative and give ranges. A CFO trusts a case that claims less and explains more.
A short example
Consider a distributor with a dozen branches. Month-end takes eleven days, rebate claims are assembled by hand each quarter, and the current ERP version loses vendor support in eighteen months. The first gate asks for a sixteen-week selection with named process owners from purchasing, sales, the warehouse and finance. The baseline records close days, rebates claimed against rebates earned, and the hours spent each month on spreadsheet reconciliations. At the second gate, the CFO compares three priced proposals against that baseline and the support deadline, and decides.
Who should present it
The initiator writes the case, but it lands better when an operating executive presents it alongside finance. When the COO, a plant manager or a branch leader stands behind the numbers, the case stops looking like an IT project. In a private-equity-owned company, show how the case connects to the value-creation plan, because that is how the operating partner will read it; our guide to ERP in the first 100 days after an acquisition covers that view.
Where an independent adviser helps
An adviser can help build the baseline and the plan for the first gate, then run the ERP software selection so the second decision rests on evidence rather than vendor claims. We do not sell software, so the answer at the second gate can be "not yet" or "upgrade what you have." If you are still deciding whom to bring in, here is how to hire an ERP selection consultant.
Frequently asked questions
How long should an ERP business case be? Short enough to read in one sitting: a one-page summary with the decision requested, and a few pages behind it for the baseline, options, plan and risks.
Should we include an ERP ROI calculation? Yes, as a range built from your own baseline rather than a vendor's calculator, with the assumptions shown. A payback estimate carries more weight at the second gate, when real costs are known.
What if the CFO wants the software cost before approving the selection? Give a clearly labelled planning range for the whole program and explain that the selection exists to replace that range with firm proposals and a like-for-like comparison.
What does a selection program cost? It depends on the number of sites and process owners and on whether implementation-partner selection is included. We quote a fixed scope after a first conversation.
Who usually signs? The CFO, or the CEO in a smaller company. In a private-equity-owned company the operating partner often joins the decision.
What if the CFO still says not yet? Ask which number would change the answer, then measure it for a quarter. A rejected case with a clear reason is a better starting point than an approved case nobody believes.
If you are writing the case for an ERP selection now, talk to John about your ERP program.
