ERP Software Selection: 7 Mistakes That Kill Projects

Updated July 2026

ERP software selection is the structured process of defining requirements, building a vendor long-list, narrowing it to a shortlist, running scripted demos and negotiating a contract. Most ERP failures are traced back to this stage — not to the implementation that follows it.

This guide covers the seven mistakes that most often derail an ERP software selection, how many vendors belong at each stage, how long the process realistically takes, and what to watch for in the contract.

ERP Software Selection Defined

ERP software selection is everything that happens between "we think we need a new system" and "we have signed with a vendor and a partner." It is deliberately separate from implementation: selection decides what you buy and on what terms, implementation decides how well you deploy it.

A complete selection covers five things:

1. Requirements — what the business actually needs the system to do, documented and prioritized.
2. Market scan — which products credibly serve your size, industry and geography.
3. Evaluation — scripted demos, RFP responses and reference calls scored against your requirements.
4. Commercials — licensing, implementation fees, and the multi-year total cost of ownership.
5. Decision and contract — the signed agreement, including renewal protection and scope definition.

Skip or rush any one of those and you inherit the consequences for the next seven to ten years. If you want the scoring dimensions in depth — functional fit, vendor viability, technology roadmap, total cost, industry references — our dedicated guide to ERP selection criteria covers how to weight and score each one. This page focuses on the process itself and the traps in it.

The 7 ERP Software Selection Mistakes That Kill Projects

Mistake 1: Starting Vendor Conversations Without Documented Requirements

The number one mistake in ERP software selection is getting requirements gathering wrong — or skipping it entirely. Between 50% and 60% of the companies we speak to have never documented what they need before they start taking vendor calls.

The consequences compound. You waste your own team's time. You waste the time of vendors and partners, who then quote against assumptions rather than facts. And you jeopardize the implementation, because nobody — including you — can say precisely what "done" looks like.

A good requirements document behaves like a contract. It sets out what you are trying to achieve and what your needs are as a business. Implementing ERP is open-heart surgery on your operating model; you would not consent to that without paperwork.

Start with our free ERP requirements gathering template, and read how to run the workshops that populate it if you have never done this before. Prioritize every line as mandatory, value-added or nice-to-have — that ranking is what makes demo scoring possible later.

Mistake 2: Letting the Vendor Run the Demo

If you send an open invitation to "show us your product," you will see a rehearsed demo of the vendor's strongest modules against a fictional company that looks nothing like yours. You will be impressed. You will also learn almost nothing.

Send a scripted demo agenda instead: five to eight of your real end-to-end processes, with your own sample data, timeboxed, with a named person from your team scoring each one. Ask every shortlisted vendor to run the same script so the outputs are comparable. Where a vendor says "that's configurable," write down who configures it, how long it takes and whether it survives an upgrade.

Mistake 3: Shortlisting on Relationships Rather Than Fit

We regularly see companies evaluating the wrong products from the outset. Perhaps the CFO knows someone at a vendor, or the incumbent finance system's supplier has an ERP product and got there first. Neither is a reason to shortlist.

Shortlist on suitability to your company size, industry and specific requirements. There is no single "best ERP" — a product that is superb for a 90-person professional services firm is a poor fit for a 900-person discrete manufacturer with three plants.

Rather than repeat a thin product roundup here, we maintain full, independent write-ups of each system, deployment model and price point. Start from our ERP vendor directory to build a credible long-list, then use the side-by-side ERP comparison to narrow it. Cross-check every candidate against your weighted criteria before it earns a demo slot.

Mistake 4: Evaluating Too Many Vendors (or Too Few)

Evaluate twelve products and you will give none of them a serious look; the process drifts, stakeholders disengage, and six months later you pick whoever chased hardest. Evaluate one, and you have no benchmark and no negotiating leverage.

The next section sets out the numbers that work.

Mistake 5: Judging the Software and Ignoring the Implementation Partner

For most mid-market buyers, the reseller or systems integrator has more influence over the outcome than the product does. The same ERP delivered by a partner with ten references in your industry and by one with none will produce two very different projects.

Evaluate them as separate decisions. Ask for the named consultants who would actually staff your project — not the pre-sales team — their industry experience, their methodology, and three references from projects of similar size completed in the last 18 months. Call all three.

Mistake 6: Budgeting the License Fee Instead of Total Cost of Ownership

License or subscription fees are usually the smallest line in an ERP budget. The full picture includes implementation services, data migration, integrations, training, third-party add-ons, internal backfill for seconded staff, ongoing support and — critically — years two through five of subscription.

Model a five-year TCO for every shortlisted option before you compare prices. A product that looks 20% cheaper in year one is frequently the most expensive over the term.

Mistake 7: Signing Without Renewal Protection

The most common commercial trap: a genuinely good deal on a one-year cloud subscription, followed twelve months later by a 30-50% increase. The remedy at that point is to rip out a system you spent nine months implementing — which is no remedy at all.

Fix it in the original contract, before you have any switching costs. Details are in the negotiation section below.

Long-List, Shortlist, Demo: How Many Vendors at Each Stage

There is a clear consensus on this, and it is one of the most useful numbers in the whole process:

Three practical rules sit behind those numbers:

• Never demo more than three products. A proper scripted demo is a half to a full day per vendor, plus scoring and internal debrief. Beyond three, your evaluators stop discriminating and start pattern-matching on presenter charisma.
• Always carry a runner-up into negotiation. Do not tell the winner they have won until the contract is signed. The credible alternative is the only real leverage you have.
• Cut the long-list on mandatory requirements only. Use "nice-to-have" items to separate finalists, never to eliminate at the long-list stage — that is how good-fit products get dropped for cosmetic reasons.

Your ERP Selection Process, Stage by Stage

Two stages are routinely underestimated. Requirements slips because process owners have day jobs — book the workshops in advance and protect them. Commercials slips because legal review is not started until the business has already emotionally committed, which destroys leverage; brief legal at stage 5, not stage 8.

If you want the evaluation mechanics in more depth — scorecards, weighting, RFP response handling — see our full ERP evaluation and selection guide, and use the ERP RFP template and process for stage 5.

Negotiating Your ERP Contract Without Getting Burned

Once you have selected a system you are not out of the woods. You still have to negotiate the software and implementation contracts, and this is where the same pitfalls recur.

ERP Software Costs and Licensing

Two traps account for most of the damage.

Year-one discounting without uplift protection. Cap annual increases in the original agreement — a fixed percentage or CPI-linked ceiling, for the full initial term and any renewal term. Ask for the renewal price in writing before you sign the first term.

Licensing the wrong thing. Companies routinely fail to license themselves correctly and fall foul of the small print. One client bought a system that had to integrate with a high-volume billing platform processing hundreds of thousands of transactions a year, and assumed the integration was free. It was not: every transaction written into the ERP required a license. Correcting it cost 20% of the planned ERP budget.

Read the metrics definitions carefully — named user versus concurrent, full versus limited/self-service users, transaction or document volumes, integration and API call allowances, sandbox and non-production environments. Get the projected volumes into the contract as an agreed baseline.

Fixed Price vs Time and Materials Contracts

Fixed price sets an expectation of an upper limit — but it can be caveated to high heaven by the implementation provider, which undoes the certainty it is supposed to provide. Read the assumptions schedule as carefully as the price. Time and materials projects can spin out of control if the right governance model is not in place on both sides: insist on a not-to-exceed cap, weekly burn reporting and a change-control process with named approvers.

Whichever you choose, document how you will pay for and contract the implementation with the same rigor you applied to the software. The ERP selection criteria scorecard should include commercial terms as a weighted dimension, not an afterthought — and if you want a ready-made version, use our vendor selection criteria checklist.

Frequently Asked Questions

How many ERP vendors should you shortlist?

Build a long-list of 6-10 vendors, narrow it to a shortlist of 3-5 for the RFP, and take only 2-3 through to scripted demos. Fewer than three at the demo stage leaves you without a benchmark or negotiating leverage; more than three and evaluators stop discriminating between products. Always keep a credible runner-up alive until the contract is signed.

What is the ERP selection process step by step?

Eight stages: mobilize the team and confirm the sponsor; document and prioritize requirements; agree a weighted scoring model; build a 6-10 vendor long-list; issue an RFP and cut to 3-5; run scripted demos with your own data and processes; complete reference calls and partner due diligence; then model five-year TCO and negotiate the contract. Requirements must be finished before any vendor conversation begins.

How long does ERP selection take?

A mid-market ERP selection typically takes 3-6 months from mobilization to signature. Enterprise or multi-entity selections run 6-12 months. Requirements gathering (3-6 weeks) and demos (4-6 weeks) are the two longest stages. The most common cause of overrun is process owners being unavailable for requirements workshops, so book and protect those sessions before the project starts.

What are the biggest ERP software selection mistakes?

The costliest are: talking to vendors before requirements are documented; letting vendors script their own demos; shortlisting on relationships instead of fit; evaluating too many products at once; ignoring the implementation partner; budgeting license fees rather than five-year total cost of ownership; and signing without a cap on renewal price increases.

How much does a bad ERP selection cost a company?

Beyond the direct 20-40% budget overruns typical of poorly scoped projects, the real cost is a seven-to-ten-year commitment to the wrong platform, plus the disruption of an eventual replacement. Undocumented licensing metrics alone have cost clients we work with around 20% of their planned ERP budget to correct after signature.