Choosing an ERP: a step-by-step guide for UK businesses

Choosing an ERP is one of the few technology decisions that shapes how a company operates for many years. Finance, purchasing, stock, production, sales and reporting all end up running through the same system, so a poor choice is felt in every department. A good choice is rarely the result of a clever product comparison. It is the result of a clear method.

This guide explains how to choose an ERP in a way that suits a UK small or mid-sized business: how to define your needs, which selection criteria matter, how to build a scoring grid, how to run demonstrations that tell you something, and how to check references before you sign. If you are still getting to grips with the basics, start with what an ERP is and come back.

Why the way you choose matters more than what you choose

Most failed ERP projects did not fail because the software was bad. They failed because the company bought a product that did not match its processes, because the scope was never agreed, or because the integrator was a poor fit for the sector. Our guide to common ERP project mistakes covers these patterns in detail, and nearly all of them can be traced back to the selection stage.

A structured ERP selection process protects you in three ways:

• It forces the leadership team to agree what the system is for before any supplier is in the room.
• It makes suppliers answer the same questions, so proposals can actually be compared.
• It leaves a written record of why a decision was taken, which is valuable when the project meets its first difficulty.

Step one: define the business case and the scope

Before you open a single vendor website, write down why you are changing system. The reason shapes everything that follows.

Typical triggers include:

• A growing number of spreadsheets bridging the gaps between an accounting package and the rest of the business.
• Stock figures that nobody trusts, or orders that cannot be tracked from quotation to delivery.
• Month-end closing that takes too long because data must be rekeyed or reconciled by hand.
• An accounting package that has reached its limits as the company adds entities, currencies or sites.
• Regulatory pressure, such as VAT reporting under Making Tax Digital or the move towards electronic invoicing.

Then decide the scope. Which functions must the new system cover, and which can stay outside it? A company that only needs better financials and stock control has a different shopping list from one that wants production planning, a warehouse module and an online shop connected in real time. Writing the scope down avoids the most common drift in a project: adding modules halfway through because they seemed attractive.

Name a project sponsor at board level and a project lead who understands the daily operation. Without both, the selection will be driven by whoever shouts loudest.

Step two: document your processes and requirements

An ERP supports processes, so you need to describe yours. This does not require a heavy consulting exercise. It requires time with the people who do the work.

For each area, ask:

1. What happens today, from trigger to outcome?
2. Where do errors, delays and duplicate entries occur?
3. What is specific to your business and cannot be compromised?
4. What would you be happy to change if the standard way of working is sound?

That last question matters. Many companies discover that a good part of what they called a "special requirement" is simply a habit. A well-chosen ERP often brings proven practice with it, and bending your process to the standard flow is cheaper and safer than bending the software.

Turn what you learn into a requirements document. Our ERP requirements template gives you a structure for this. Classify each requirement as essential, important or desirable. Essential means that without it the system cannot go live; desirable means it would be pleasant but nobody will walk away over it.

Step three: understand the options before you shortlist

You do not need to know every product on the market. You need to understand the main families of solution and which one corresponds to your size and complexity.

The five solutions covered on this site illustrate the range: Odoo, Microsoft Dynamics 365 Business Central, Sage, Oracle NetSuite and SAP Business One. We are not affiliated with any of these vendors, and the point is not to push you towards one. Each suits a particular profile. For side-by-side reading, see our ERP comparisons, including the best ERP for a small business.

At this stage, aim for a shortlist small enough to examine properly. A long list of products produces shallow evaluation; a short list produces real answers.

Step four: set your selection criteria

Selection criteria should be agreed before you see any demonstration, otherwise the most polished presentation sets the standard. Group them into five families.

Functional fit

Does the standard product cover your essential processes without heavy customisation? Check the areas that actually drive your business: financial management and VAT, purchasing, sales order management, inventory and warehousing, production or project management, reporting. Sector-specific needs deserve extra attention. A manufacturer, a distributor and a professional services firm need very different things, which is why our pages on manufacturing ERP, wholesale and distribution ERP and professional services ERP exist.

Look closely at UK-specific points: VAT handling, Making Tax Digital compatibility, bank feeds, payroll connections, multi-currency with sterling as the base currency and the ability to deal with electronic invoicing. Our guide on ERP and e-invoicing explains what to look for.

Technical fit

How will the system sit within your existing landscape? Consider:

• Integration with your e-commerce platform, CRM, payroll, banking and any specialist systems.
• Availability of open interfaces and the way integrations are built and maintained.
• Data migration: the quality of your current data and how much clean-up is needed.
• Security, access control and compliance with UK GDPR, including where data is hosted.
• Hosting model: cloud, hybrid or on-premise. Our guide to cloud versus on-premise ERP helps you weigh this.
• Performance and the capacity to grow with transaction volumes and user numbers.

Total cost of ownership

A licence or subscription price is only one line of the budget. A fair comparison looks at the whole cost over the period you expect to use the system: software, implementation services, data migration, integration work, training, support, hosting where relevant, upgrades and the internal time your team will spend. Ask every supplier for a written breakdown on the same basis. Our guide to ERP cost shows how the pieces fit together and where budgets typically slip.

Be cautious of proposals that look cheap because the scope is thin. A low quotation that excludes migration, integrations or training simply moves the cost to a later date.

Integrator ecosystem

For most mid-sized companies, the integrator matters as much as the product. The same software can deliver a smooth project or a painful one depending on who configures it. Evaluate:

• Whether integrators with real experience of your sector are available and willing to take your project.
• The depth of their team: functional consultants, developers, project managers and support.
• Their method for scoping, testing, training and handover.
• How they handle support after go-live, and what happens if a key consultant leaves.

A product with a large and competitive partner network gives you negotiating power and choice. A product with only a few integrators in your region may leave you dependent on one supplier. Our guide to choosing an ERP integrator goes deeper on this half of the decision.

Vendor viability and product roadmap

You are choosing a platform that you expect to use for years. Consider the financial standing of the vendor, the clarity of its roadmap, how it handles upgrades, and whether it continues to invest in the product line you are buying. Be wary of a product being retired or sidelined in favour of a different offering. Neutral sources, such as published vendor announcements and independent analyst commentary, are better evidence than a sales conversation.

Step five: build a weighted scoring grid

A scoring grid converts a subjective debate into a structured one. It does not make the decision for you, but it makes the reasoning visible and exposes disagreement early.

The method is simple. List your criteria in rows. Give each a weight reflecting its importance to your business. Then score every shortlisted solution, and every integrator proposal, against each criterion, using the same scale. A simple scale works best: for example, weak, adequate, strong, with a numerical equivalent agreed by your team before scoring begins.

Here is an example layout:

A few rules keep the grid honest:

• Fix the weights before scoring. Otherwise people adjust weights to favour the product they already like.
• Have several people score independently, then compare. Large differences in scoring reveal different assumptions, which are worth discussing.
• Score evidence, not impressions. A feature seen in a scripted demonstration earns a score; a feature promised for a future release does not.
• Treat essential requirements as gates. If a solution cannot meet an essential requirement, no total score should rescue it.

The final ranking is an input to your decision, not a verdict. If the result surprises you, ask why. Sometimes the grid reveals that a criterion was mis-weighted; sometimes it exposes a bias.

Step six: run scripted demonstrations

A standard demonstration shows a product at its best, using clean sample data and a path chosen by the supplier. It tells you very little about how the product will behave in your business. Insist on a scripted demonstration instead.

Prepare the script in advance and send it to each supplier so that all of them respond to the same scenarios. Good scenarios come from your real processes:

• Take a customer order from quotation to invoice and payment, including a partial delivery and a credit note.
• Process a purchase from requisition to goods receipt and supplier invoice matching, including a price difference.
• Run a month-end close and produce the management reports your board actually reads.
• Handle a real exception: a returned product, a back-to-back order, a stock adjustment, or a customer with special pricing.
• Show how a user without technical knowledge finds an answer in the system, rather than asking IT.

Ask the people who will use the system daily to attend, not only managers. They are the ones who will tell you whether the screens are workable. Note that a demonstration also lets you see the integrator at work: how well they listen, how honestly they say "that requires customisation", and whether the people presenting are the people who will deliver.

Where a supplier says a requirement is possible, ask whether it is standard, configured or custom-developed. The answer has consequences for cost, upgrades and risk.

Step seven: check references

References are the most underused part of ERP selection. A supplier will gladly name satisfied customers; your job is to make the conversation useful.

Ask for references that resemble you in sector, size and complexity, and ideally some where the project was delivered by the same integrator team you are considering. When you speak to them, ask open questions:

• What did the project look like in practice compared with what was promised?
• What was harder than expected, and how did the integrator respond?
• How did the team handle data migration, testing and training?
• What does support look like now, a while after go-live?
• If you were starting again, what would you change?

Listen as much to what is not said as to what is. A reference who hesitates when asked about support or change management is giving you information.

We never name integrators or clients on this site, but when you compare real proposals you are entitled to ask each supplier for these references. Treat a reluctance to provide them as a signal in itself.

Step eight: evaluate the proposal and the contract

Once demonstrations and references have narrowed the field, compare the formal proposals line by line. A good proposal is specific about scope, assumptions, responsibilities, deliverables and what is excluded.

Check in particular:

• Scope and assumptions. Is it clear which processes, entities and integrations are included? Are the assumptions realistic for your business?
• Project governance. Who is the project manager? How are changes to scope handled and priced?
• Your own commitments. Every ERP project depends on your team's time. A proposal that ignores this is optimistic.
• Data migration and testing. Is there a clear approach, and who is accountable for data quality?
• Training and go-live support. Is support planned for the critical period after launch?
• Licensing and exit terms. How does pricing change as you add users, entities or modules? What happens to your data if you leave?

For a sense of how the project unfolds after the decision, read our guide to ERP implementation, and if you are replacing an existing system, the guide to ERP migration.

Common selection mistakes to avoid

Certain errors recur often enough to be worth naming.

• Starting with the product rather than the process. Companies fall in love with a demonstration and then try to justify it.
• Letting IT or finance decide alone. An ERP spans the whole business. Operations, sales, purchasing and warehouse staff must be heard.
• Ignoring the integrator. A good product with a weak delivery team is a poor investment.
• Underestimating change management. Users who were not involved will resist a system they did not choose.
• Choosing on price alone. The lowest quotation is not the lowest cost once scope gaps appear.
• Over-customising. Every bespoke feature must be maintained and retested at each upgrade. Reserve customisation for what truly differentiates you.
• Rushing the decision, or dragging it out. A rushed process skips the needs analysis; a drawn-out process loses momentum and sponsor attention.

Choosing an ERP for a small business

Smaller companies often assume that a structured selection is only for large organisations. In fact, the discipline matters just as much, because a small business has less room to absorb a failed project. The process can be lighter: a shorter requirements document, a tighter shortlist, a single round of scripted demonstrations. But the logic is the same. Our guide to ERP for small business explains how to scale the method to your size.

Sector matters

The more specific your sector, the more the selection should concentrate on proven fit rather than on general features. A food producer needs traceability and batch management; a contractor needs job costing and subcontractor handling; an online retailer needs fast order flows and channel integration. Our sector pages, such as food and beverage ERP, construction ERP and e-commerce ERP, set out what to examine in each case. Ask integrators to show a project in your sector, not a generic one.

How a matching service fits into your selection

Once your needs are written down, the slow part is finding integrators who genuinely suit your project and are available to take it on. That is where a free matching service helps. ERP Integrator puts you in touch with an integrator suited to your project, so that you can compare real proposals using the grid above. The integrator carries out the project; we make the introduction. We are neutral between vendors and do not favour any product.

A short checklist to finish

Before you sign, confirm that you can answer yes to each of these:

• Has the leadership team agreed the business case and the scope?
• Are the requirements written down and ranked by importance?
• Were criteria and weights fixed before the demonstrations?
• Did every supplier respond to the same scripted scenarios?
• Have you spoken to references resembling your company?
• Is the total cost of ownership compared on a like-for-like basis?
• Are the roles of your own team and the integrator clearly defined in the contract?

If the answer to any of these is no, you have identified your next piece of work, and it is better done now than after the contract is signed.

Next steps

Start with your processes, write your requirements, agree your criteria and score with discipline. Then use real proposals, not marketing material, to make the final call. If you would like help finding an integrator for a project of your size and sector, you can describe your project and we will introduce a suitable integrator at no charge.

Software names mentioned are trademarks of their respective publishers and are referred to for information only. This site is independent and is not affiliated with any software publisher.