What to Expect During ERP Discovery

The short answer

Discovery is the stage where the team learns how your business really runs, before deciding what to build. It ends with a written description of your process and a proposed scope that you have confirmed. If you are asked to sign off on a build before that exists, push back.

What does discovery cover?

A good discovery looks at six things.

• Workflows: how an order, a purchase, a hire or a payment actually moves through your business, step by step, as it happens today and not as the manual says.
• People and roles: who does each step, who approves it, and who only needs to see it.
• Data: where the information lives today, in what shape, and how clean it is. Spreadsheets with text typed into number columns are common and need to be dealt with.
• Bottlenecks: where work waits, gets re-typed or goes wrong. These are usually where the software earns its cost.
• Integrations: what the new system needs to exchange data with, for example accounting software or an existing ERP.
• Reporting: the numbers management actually reads, and where they come from.

Who needs to be involved?

The people who do the work, not only the people who sign the contract. A manager's description of a process often differs from what the person doing it describes. Plan for the time of a few of the actual users.

What should you have when discovery ends?

You should receive four things in writing.

• A map of your current process, which you read and correct.
• A scope: what will be built, in what order, and what is deliberately left out.
• An outline of the system's structure, covering roles, permissions and integrations.
• An honest view on whether a build is the right answer. Sometimes the answer is a smaller tool or a change in process.

Why is it worth the time?

It is cheaper to correct a process map than to rebuild a system. Most of the mistakes that become expensive later are misunderstandings that a few conversations would have caught.

What should you ask before you start?

Three questions are worth asking any team before discovery begins.

• Who will do the discovery, and will the same people build it?
• What will I have in writing at the end?
• What happens if discovery shows a build is not the right answer?