ERP readiness checklist: twelve questions before you buy anything

An ERP system does not fix a business. It makes the way the business already works faster and more visible — including the parts that do not work. Most of what decides whether a project succeeds is settled before any software is bought.

These twelve questions take an afternoon. Answer them with the people who do the work, not only with management, and write the answers down. There are no right answers, only honest ones.

Your processes and the people who run them

• Are your main processes written down — from a customer order to cash in the bank, and from a purchase request to a supplier payment?

  If a process lives only in people's heads, the system will be configured around whoever explains it loudest.

• Does each process have one named owner who can say how it should work?

  Configuration is a long series of small decisions. Without an owner, each one waits for a meeting.

• Who inside the business will own the system once it is live?

  Someone has to add users, answer daily questions and notice when people drift back to spreadsheets. A supplier can support that person but cannot be that person.

• Can the people who will use it every day give time to training and testing — during working hours?

  Training squeezed in after a full day's work does not stick, and the old habits come back.

Your data

• Is there one item list, with one code per item and no duplicates?

  Two codes for the same product means two stock figures, and neither is right.

• Are your customer and supplier lists current, with balances you would be willing to show them?

  A balance you cannot defend to the customer should not be loaded into a new system as if it were fact.

• Can you produce opening balances — bank, stock, receivables, payables — as of one date, agreed with your accountant?

  Everything the system reports afterwards is built on these figures. If they are wrong, every report is wrong from the first day.

• Have you decided how much history to bring across, and what stays in the old files?

  Migrating years of old transactions is expensive, and they are rarely opened again. Open balances and open documents are usually enough; the rest can stay archived.

The decision

• Can you state the problem you are solving in one sentence, without naming any software?

  'We need an ERP' is a purchase, not a problem. 'We do not know our stock until we count it' is a problem.

• What happens if you do nothing for another year?

  If the honest answer is 'not much', the project will lose every argument about priorities once it starts.

• Is a full ERP the right size of tool for that problem?

  A single shop or a small restaurant often needs a good point of sale and clean books, not a company-wide system. Buying too much is a mistake, just as buying too little is.

• How will you know, a year after go-live, that it worked?

  Pick two or three things you can measure today — days to close the month, stock differences at count, time to issue a quotation — and write down where they stand now.

The questions you cannot answer are the work that has to happen before configuration — whoever ends up doing it.

Take the written answers to any supplier you speak to. A good one will read them before showing you a demonstration, and will tell you which gaps they can help close and which are yours. If most of the answers are 'not yet', that is not a reason to drop the idea. It is a reason to start with the item list, the balances and the process owners — work worth doing whether or not you ever buy a system.