Most companies start their ERP selection process backwards. Someone gets frustrated with the current system, starts Googling alternatives, and within a week has three vendor demos booked — before anyone has actually written down what the business needs the new system to do. Demos are persuasive by design, and comparing systems without your own requirements defined means you're evaluating against the vendor's pitch, not against your business.
The fix isn't complicated, but it does require discipline most companies skip under time pressure. Here's a five-step process that puts requirements first.
Step 1: Document What's Actually Broken, Specifically
Not "our system is outdated" — specifically, what breaks, how often, and what it costs. "Month-end close takes 12 days because inventory reconciliation is manual" is a requirement. "We need something more modern" is not. Spend a week having each department — finance, operations, sales — write down the three most time-consuming manual workarounds they currently live with. This becomes your actual requirements list, grounded in real pain rather than vendor feature lists.
Step 2: Separate Must-Haves From Nice-to-Haves
Once you have a real list, sort it ruthlessly. A must-have is something that, if missing, makes a system unusable for your business — multi-location inventory tracking for a distributor, job costing for a construction firm, multi-currency consolidation for a company with a US subsidiary. A nice-to-have is something that would help but that you're currently working around acceptably.
This step matters because vendor demos are built to showcase everything a system can do, and it's easy to get pulled toward features that sound impressive but don't address your actual must-have list. Keep the list short and specific — five to eight genuine must-haves is normal; twenty is usually a sign the list hasn't been prioritized yet.
Step 3: Set a Realistic Budget Range Before Shopping
ERP implementation cost varies enormously based on complexity, and going into vendor conversations without a budget range means every proposal looks either surprisingly cheap or shockingly expensive with no context. A rough range — informed by your company size and the complexity documented in Step 1 — gives you a filter before you're emotionally invested in a specific vendor's pitch.
Budget should account for more than software licensing: implementation services, data migration, training, and a reasonable contingency for scope that gets discovered mid-project. A budget that only covers licensing cost is not a real budget for this kind of purchase.
Step 4: Shortlist Against Requirements, Not Reputation
With a must-have list and a budget range in hand, shortlist three to four systems that plausibly meet your specific requirements — not the three most-mentioned names in generic "best ERP" articles. A system with a strong general reputation that doesn't handle your specific inventory or job-costing needs isn't a good shortlist candidate regardless of its market position.
This is also the point to start evaluating implementation partners, not just software. The same ERP platform can produce very different outcomes depending on who implements it — a topic worth its own separate evaluation, covered elsewhere in this series.
Step 5: Demo Against Your Own Scenarios, Not the Vendor's Script
Every vendor demo has a default script built to show the system at its best. Instead of accepting that script passively, bring your own specific scenarios from Step 1 — the actual manual workaround that's costing you time — and ask the vendor to show you exactly how their system handles it. A vendor who can walk through your specific scenario confidently is telling you something real. A vendor who redirects to a different, more flattering example is telling you something too.
What Happens When Companies Skip This
Skipping requirements-first selection doesn't always produce a visibly bad outcome — sometimes a company gets lucky and the system happens to fit anyway. More often, it produces a system that handles 70% of operations well and requires ongoing manual workarounds for the remaining 30%, because nobody checked those specific requirements before signing. Those workarounds tend to become permanent, quietly reducing the value of the whole investment for years afterward.
Where Business Central Fits This Process
Business Central tends to perform well against this kind of requirements-first evaluation for companies with genuine inventory, job costing, or multi-entity complexity — its strength is depth in exactly those areas. It's a legitimately poor match for a company whose must-have list is short and simple; that mismatch shows up quickly once the requirements exercise above is done honestly, which is the entire point of doing it before shopping rather than after.
The Practical Takeaway
Running this process takes real time upfront — usually two to four weeks of internal work before a single vendor conversation. That investment consistently pays for itself by preventing the far more expensive mistake of selecting a system that looked impressive in a demo but doesn't actually fit the specific, documented reality of how your business operates.