This week’s Featured Blog Friday comes from our Senior AGR Consultant, Alan Iles, who works in our UK office. As always, if you have any questions or comments regarding this blog post, feel free to comment on this blog post, tweet us @AGRDynamics, or contact us here.

A guest speaker at our user conference earlier this year gave a fantastic presentation about his experience implementing a new ERP system. The stand out lesson/message that I took away from that talk was, as you might have guessed, the importance of data in the process.

I think everyone understands how important data is, especially in our increasingly technological world. We all know that the prices on the system must be correct,, we must know what our stock levels are, etc. We need to know what we’ve sold, what we’ve paid for, who owes us what – the list is almost endless. What is not understood so often, or perhaps more accurately, what is underestimated, is the difficulty in managing that data:

  • Collecting it: Where is it all? Existing systems, emails, pdf files, spreadsheets, people’s heads, on paper, with the suppliers? In fact, does it even exist?
  • Loading it: This one can be a nightmare. The kind of databases that most of our ERP systems are on are very particular. Everything must be just so. Get it wrong and you will be there for hours, days, maybe even weeks. But get it right and you can load thousands, millions even, of rows of data in a very short space of time.
  • Maintaining it: Who updates it? How often is it updated? How do you know it is correct? When you have 100,000 live items across 70 locations, that’s 7 million rows of data. That’s a lot of data to review and update. If you change 50% of your product range every year, it’s a lot of work to consider.
  • Using it: Is the data available? Is it in the right format (see Loading it above)? Is it right and up to date?

At my last company, I wanted to use IT everywhere, I wanted to optimise everything, make it all faster, automate everything, have everything centralised and available to all. Paper, spreadsheets and calculators should be banned. Everything could and should be done in the ERP and connected systems. One of the areas where I thought there could be big improvements by using our systems was in production and manufacturing. Use the software to plan the supply of components and the scheduling of the production. Seemed simple enough, bill of materials, demand, some quick sums and there’s your list of what to buy and manufacture and when to do it. I approached the Head of Development about this, her response was something along the lines of, “nice idea, but…”, she went on to say “…you will spend more time finding the master data, getting it loaded, validated and then keeping it correct than it takes me and my team to do it manually in excel with the knowledge in our heads”. Nonsense, I thought…

But she was right.

Well, at that point in time she was right.

She understood that at that stage in the life of the company, it was more efficient to keep it off system, to not automate. And she had the added benefit of knowing which suppliers she could bend the rules with – she could react and respond and communicate in ways that our systems cannot yet do. IT systems need rules, without them, they don’t know what to do.

She also understood the impact of the data being wrong. If any of that data was wrong or missing, the output from the system would be at best, worthless, at worst, a liability. The old adage “crap in, crap out.” No matter how good your system, whatever its purpose, if it hasn’t got the data it needs, it is meaningless.

There comes a time though when having the only repository of critical knowledge in one or two people’s heads, or on a couple of spreadsheets, is dangerous. There comes a point when the sheer volume of products and variables makes it impossible to reliably work through using standard spreadsheets. It is at this point that there needs to be the investment in time to get that information held centrally; so it is readily available to anyone who requires it, it is backed up, and can be relied on. Once it is there, it can then also be used to improve and standardise many processes. This company is obviously now an AGR customer, but they, like many others, struggled to get the data complete and right.

Our statistical forecasting and planning engines are at the heart of AGR’s suite of tools. Both exist for the sole purpose of creating forecasts, whether they be statistical forecasts or forecasts entered by the experts running the system. Forecasts are though, by their very definition, inaccurate, it’s just a question of how inaccurate. Improving the quality of the data fed in directly improves the quality of the forecast out, that could be data fed to the software or to the Planner.

The AGR Inventory Optimiser does not deal in inaccuracies. It needs to know the supplier, not a guess at the supplier; it needs to know the current stock, not an estimate; it needs to know the rules for buying: the minimum order quantities, the batch sizes, etc. It is here that it becomes critical that everything is exactly right.

AGR integrates directly with most ERP systems to ensure it is using the most accurate and up to date data available, be that sales information, purchasing information, or replenishment routes. So at least you only need to get/keep it right in one place.

In summary, don’t underestimate the difficulty data can present. Especially to anyone looking to implement a new ERP system. You cannot go live without the data so make sure it is right plenty of time before you plan to start testing the actual system.