ERP Evaluation Criteria: What to Score Vendors On

Updated September 2026.

ERP selection fails differently than most software decisions. A CRM that doesn't fit gets swapped out in six months with limited damage. An ERP that doesn't fit gets customized around for three years, breaks on every version upgrade, and takes the finance close down with it when it finally gives out.

Most evaluation templates hand a committee the same categories used for any SaaS purchase: features, integrations, price, support. That misses what actually separates a good ERP decision from an expensive mistake: whether the data model matches how your entities really work, whether the implementation partner can execute, and whether what you buy today survives the version after next. Our full ERP buying guide covers the end-to-end process; this piece is the scoring checklist to run during evaluation itself.

Key Facts: ERP evaluation

• Panorama Consulting's 2026 ERP Report (170 organizations surveyed, median annual revenue $200.5 million) found a median project timeline of 9 months, with more than a quarter of projects still going over budget and almost a quarter going over schedule.
• Of the projects that went over budget, the most common cause was an unexpected need for additional technology, not scope creep from the buyer. Of the ones that went over schedule, the most common cause was internal organizational resistance, not the software itself (Panorama Consulting, 2026 ERP Report).
• Microsoft openly publishes Business Central list pricing at $80/user/month for Essentials and $110/user/month for Premium, both paid yearly (Microsoft), despite the tier being routinely described as "quote-based" in buyer content. Published list pricing is the exception in this category, not the rule, which is why per-user comparison shopping breaks down here.

What makes ERP evaluation criteria different

A CRM evaluation asks whether reps will use it. An accounting software evaluation asks whether the ledger closes cleanly. An ERP evaluation asks both at once, across every department, entity, and currency, plus whether the data model can hold your business structure without a workaround.

Generic vendor scorecards undersell this risk. A general SaaS vendor evaluation treats implementation as a line item. In ERP, implementation quality often matters more than the software: the same platform can be a smooth 6-month rollout with one partner and an 18-month recovery project with another. Score the partner as rigorously as the platform.

The ERP evaluation criteria checklist

Score each criterion 1-5 during demos, reference calls, and (ideally) a paid discovery workshop with your shortlist. The columns below describe what a strong answer looks like and what should knock a vendor down a tier.

Suite vs. best-of-breed: decide this before you demo anything

This is the fork most CRM-style checklists never have to make, and it changes which criteria matter most. Decide it early, before you waste time on demos that were never going to fit.

How to weight the criteria by buying situation

The right weighting depends on your entity structure and regulatory load more than headcount alone.

Single-entity buyers weight module fit highest because they're solving one clear operational problem. Multi-entity teams shift weight to the data model, since that's what breaks first when a second subsidiary or currency shows up. Enterprises weight change management heaviest because at that scale, the technical implementation is rarely the reason a project fails; the organization's ability to absorb the change is.

The data model test most buyers skip

Ask every finalist to walk through your actual entity structure, not a demo company. If you have, or plan to have, more than one legal entity, currency, or intercompany transaction, this single test disqualifies more vendors than any feature comparison.

Panorama Consulting groups ERP platforms into tiers by the complexity of business they're built to handle, a more useful lens than "SMB vs. enterprise" alone:

Source: Panorama Consulting, 2026 ERP Report. Business Central, the SMB-focused product most small and mid-size buyers actually price out, sits below Dynamics 365 Finance and Supply Chain here; don't assume every product under a big vendor's umbrella targets the same tier.

Why the implementation partner matters as much as the software

In most software categories, the vendor and the implementer are the same conversation. In ERP, they're often not, and the partner is frequently the bigger risk.

Customization vs. configuration, and what it does to your upgrade path

Every custom line of code is a future upgrade cost. Configuration (fields, approval workflows, role permissions) generally survives a vendor's version updates; custom code built against a specific version frequently does not. Vendors are increasingly forcing this choice for you: Epicor recently announced that all new features for its Kinetic, Prophet 21, and BisTrack products go to the cloud exclusively, putting heavily customized on-premises deployments of those products on a fixed clock (Panorama Consulting, 2026 ERP Report).

Ask every finalist, module by module, which requirements need custom code, and what happens to that code on the next major version. A vague answer previews an unbudgeted redo down the line.

Phased vs. big-bang: how to choose your rollout

This decision belongs in the evaluation phase, not after signature, because it affects which vendors and partners can realistically support it.

More than a quarter of organizations now use a hybrid approach rather than a purely phased or purely big-bang rollout, making it the single most common pattern among multi-entity and multinational implementations, per Panorama's 2026 ERP Report. It wins because it protects revenue-critical operations with an early go-live while giving later entities a slower, lower-risk runway.

Total cost of ownership: what actually gets billed

License is usually the smallest line on the invoice. Model all four layers before comparing vendors on sticker price alone, or you'll rank finalists on the least meaningful number in the deal.

Where the category is genuinely quote-only, and where it isn't. Most mid-market and enterprise ERP still hides behind "contact sales," and that opacity is itself worth naming. But two vendors often assumed to be quote-based actually publish list prices:

Two things worth knowing if you have seen different numbers elsewhere. Odoo's free tier ("One App Free") covers exactly one app with unlimited users, not the full suite, so a "free full ERP" claim is a misread of that plan. And Business Central is not quote-based: its top published tier (Premium, $110/user/month paid yearly) runs higher than the "$70-100/user/month" range often repeated in buyer content.

Why so many ERP projects overrun

The failure pattern is more predictable than most buyers expect, which means most of it is preventable during evaluation, not something you discover mid-project. Panorama's 2026 data shows more than a quarter of projects going over budget and almost a quarter going over schedule, and the causes are specific, not vague "ERP is hard" hand-waving.

Budget overruns most often trace back to an unexpected need for additional technology: a system-fit gap surfaced late and got patched with extra tooling instead of caught during selection. Schedule overruns most often trace back to organizational resistance: delayed sign-offs and resistance to standardizing a chart of accounts, not technical delays. Change management is what buyers underinvest in most: fewer than a quarter report an intense focus on it, even though it's the top named cause of schedule slips (Panorama Consulting, 2026 ERP Report). The same report found post-selection services, change management and benefits realization especially, growing faster year over year than pure selection support: the scorecard gets you to a good contract, but what happens after signature is where the project is actually won or lost.

Key questions to ask vendors and their implementation partners

Ask these after the demo, when the sales engineer is off script and the partner's delivery lead is in the room.

1. "Walk me through how a new subsidiary or mid-year acquisition would flow through your data model." Exposes a data model that only looks multi-entity in the slide deck.
2. "Which requirements need custom code versus configuration, and what happens to that code on your next major version?" Get it in writing, module by module.
3. "Introduce us to a same-industry, same-revenue-band client who went live in the last 18 months, without you on the call." A vendor who hesitates is protecting you from hearing something.
4. "What's included in your implementation fee, and what triggers a change order?" Ambiguity here is where most budget overruns start.
5. "What does your average project look like: on time and on budget, or not, and why?" A partner who can't answer honestly hasn't tracked their own delivery record.
6. "If we skip a module today, what does it cost to add later: re-implementation or just activation?" Tells you whether pricing punishes phased adoption.
7. "What happens to our data and configuration if we terminate the contract, or if you get acquired?" Get the exit path in writing before you need it.

Our SaaS vendor evaluation scorecard and software shortlisting framework cover the diligence questions that apply beyond ERP.

Frequently asked questions

How is ERP evaluation different from general SaaS evaluation criteria?

General SaaS evaluation asks whether a tool does the job, integrates with what you have, and fits the budget. ERP evaluation adds two layers: whether the data model represents your legal and operational structure, and whether the implementation partner can execute. A great platform with a weak partner routinely underperforms a decent platform with a strong one, rare in lighter categories like help desk software.

How many ERP vendors should we shortlist?

Two to three finalists is realistic. ERP demos and discovery workshops pull real time from finance, operations, and IT staff who also have day jobs, so a longer list just drags the process out. Use the criteria table above to narrow it first.

Should we pick the software or the implementation partner first?

Pick software finalists first, then evaluate partners for each, since partner quality varies within the same platform. Some platforms have one strong regional partner and several weak ones; picking the partner first risks anchoring on the best sales team rather than the best data model fit.

Is it ever fine to buy a full suite and only use one module?

Sometimes, if you genuinely expect to grow into the other modules within a reasonable window. But without a credible plan to use more than one module within two to three years, a focused best-of-breed tool (see how to choose accounting software or inventory management software) is usually the better, cheaper buy.

How do we know if our project is at risk of becoming an overrun statistic?

Watch for two leading indicators: any sign that a "must-have" capability wasn't caught during selection and needs bolt-on technology, and any resistance from process owners to standardizing how work gets done, like a shared chart of accounts or approval flow. Both show up early, often in the first design workshop, if you're watching for them instead of assuming the vendor will flag them.

Score the partner as hard as the platform

Every criterion above matters, but if you only have time to do two things well: pressure-test the data model against your real entity structure, and reference-check the implementation partner as hard as the software. Most ERP projects that go sideways weren't sunk by a missing feature. They were sunk by a data model that didn't fit or a partner that couldn't deliver, both visible during evaluation if you're scoring for them.

Related reading

• How to choose ERP software: a buyer's guide
• How to choose ERP software for small business
• SaaS vendor evaluation scorecard
• How to choose accounting software
• Accounting software evaluation criteria
• How to choose inventory management software
• CRM evaluation criteria checklist
• Best NetSuite alternatives: full ERP roundup