You tried an ERP with your own data. Your scenarios were run on screen. Some went through cleanly, and for others you heard "that would need to be set up". You left the meeting and everyone remembered something different: the sales lead liked the quote screen, accounting was not sure about the exchange-difference entry, and the warehouse lead meant to ask about handheld terminals and forgot.
This is where the value of a proof of concept is decided. If the results are not written down, the trial turns into an impression, and the decision goes to whoever speaks last. This article explains how to read the results.
Turn impressions into records
The day after the meeting, ask everyone who attended to fill in the same table. One row per scenario:
| Scenario | Result | Could our own staff do it | Note |
|---|---|---|---|
| Part shipment and invoice | Worked / Needs adapting / Did not work | Yes / No | |
| Foreign-currency purchase and exchange difference | Worked / Needs adapting / Did not work | Yes / No | |
| Stock count difference | Worked / Needs adapting / Did not work | Yes / No |
Put the tables side by side and two things show up. The jobs everyone marked as "worked" drop out of the decision. The jobs one person marked "worked" and another "did not work" go back to the seller as questions.
Have "needs adapting" spelled out
This is the answer you hear most in a trial, and it can mean three different things:
- Configuration. The software already has it and it needs to be switched on for the way you work. It is a normal part of setup.
- A change in how you work. The software does the job, but not by the method you use today. You will change your way of working.
- Development. The software does not have it and it has to be written.
Ask the seller in writing which of the three each "adaptation" is. Every job that needs development should appear as a separate line in the quote. For jobs that need a change in how you work, talk to the department concerned: is the new method acceptable?
For a subsidiary, add a fourth question here. Does head office accept the local method, or does the group process have to be reproduced exactly?
Look at the data findings separately
Problems that came up while your data was being imported are problems with the data, not with the software. They still matter for the decision, because you will have to do this work whichever software you choose.
Ask the seller what they ran into during the import:
- Which fields were missing?
- How many records were duplicated?
- Which units and codes did not match?
- What could not be imported or had to be fixed by hand?
That list is the work plan for the data clean-up before the transition. A seller who runs the import smoothly and reports the problems openly will behave the same way during the transition.
How did it sit in the user's hands?
The job the seller showed and the job your employee did are not the same thing. If you handed the mouse to your own staff in the trial, look at these:
- Could they finish the job without help?
- How many screens did they go through?
- Where did they get stuck?
It is not enough that a job can be done in the software. If a record made a hundred times a day in the warehouse takes six screens, after a while that record stops being made.
Note what was not seen
Every trial has a narrow scope and some jobs stay outside it. When you decide, do not count what was not seen as seen. For each job left out, ask the seller two questions: does the software have it, and if so, can we see that with our own data too?
Month-end closing, sending e-Fatura (e-invoices) and the screens your CPA will use are often skipped in the first trial. If they matter to you, ask for a second session.
Three questions for the decision meeting
Once the tables are in, look at three questions in the decision meeting:
- Did the three most important flows of our business work with our own data? If not, the other advantages count for less.
- Which jobs need development, and are they in the quote? A job that is not written down is a cost that will appear later.
- What do we have to do to move our data, and who will do it?
If the answers to these three are clear, so is the decision. If they are not, running the missing scenario again is better than starting setup on an unclear decision.
What happens after a trial at Birasyo?
At Birasyo we run the trial with your data, and in the meeting we note together with you what is missing or needs adapting. If you decide to go ahead, the settings and the import layout from the trial are used in the setup. If you do not, we delete the data we loaded.
To plan a trial with your own data, fill in the form and we will call you the same day.
Related reads
Share this post on LinkedIn
The title, summary and hashtags are copied to your clipboard and the sharing window opens. Paste (Cmd/Ctrl+V) and post.
