If you’ve gotten things running, built a prototype, and you’re planning to scale things up, you might have heard you need an ERP to get things really moving.

But what is an ERP? Let’s take a look at Oracle’s definition:

ERP stands for enterprise resource planning. It's a software system that includes all the tools and processes required to run a successful company.

That might be the worst and least descriptive definition I’ve ever seen of anything… they’re clearly selling a dream to distract from everyone’s nightmarish experience.

Other folks say it’s the everything system, but that needs a definition. An everything bagel has 5 ingredients, no one’s putting caviar and foie gras because we all know intuitively that “everything” isn’t a real definition.

So what is an ERP really? It’s a glorified cash register with a lot of bells and whistles. 

It transacts parts throughout your business, accounting the movement and allocation of all of the costs and inventory for those parts.

That’s its core functionality. Full stop.

It’s a ledger that tracks incoming parts through receiving to warehousing to production processes to outgoing sales.

It might do some other things… it records your employees and what they’re working on, if you want it to. It’s going to be where you manage shipping product to your customers, so it’s convenient to add some names, phone numbers, and addresses. Because it tracks receiving of parts it often includes a basic quality inspection process. But at its core, it’s a cash register. Tracking the movement of parts through various “Orders” and accounting for their costs appropriately.

It needs to know enough about the product to build the product from raw materials, so it needs a full BOM. It needs to also order the parts, so it needs to know what a “3.5 inch gold plated left filange” is, where it comes from, how much it costs, and how to get it into your warehouse.

And it’s an absolute b%^ch to set up because of what goes into doing that well. You’re essentially recreating a digital twin of your company’s operations department.

The first step therefore, is figuring out where your ERP starts and stops to prevent it spiraling into a 10-year clustermess, as your events team starts to customize it to include every employee’s preferred Chipotle order for an automated Cinco de Mayo Luncheon.

Minimizing the amount of things that go into your ERP is key to getting it operating quickly, keeping data accurate, minimizing snafus in the rollout process, and allowing for business flexibility afterwards. If the process in question is something that 10,000 other companies in the US do on a daily basis, and it relates to parts movements and cost accounting, it probably goes into the ERP.

If any piece of your business is special, part of your secret sauce or a non-standard operation, or doesn’t involve parts consumption, a good rule of thumb is to just keep it out. Finance and Warehouse people should be the defenders of the ERP vision, the ones who care that all your costs and part counts are accurate. 

The project champion needs to be more of a gatekeeper, shouting “You shall not pass” when a Balrog of an engineer tries to implement a $30,000 MES for your 1 manufacturing oven or a PLM for your 5 part BOMs. Your NetSuite consultants will try to bend over backwards to achieve every request and upsell you at every turn, you need someone who forces the default answer to be no for every request.

Most ERPs have a basic v0.1 functionality of every part of the factory software stack. They have a basic WMS, QMS, MES, MRP, CRM, VRM, PLM and any other ABC system you might need, but it’s going to be terrible at all of them out of the box, aside from its core functionality of tracking parts costs and movements.

It’s tempting to want to improve all of them out of the box, but changing 100 things at once is enough of a challenge, and trying to change 101 things at once is going to be even harder despite what the salesperson on the other side of that Zoom call is telling you. 

Define explicitly what you don’t need the ERP to do and stick to it. Get a basic functionality set up working before you go in adding bells and whistles everywhere. Make sure your master data is clean and accurate, and get the fundamentals working well. 

Naturally some things are needed on day 1 so don’t take this to an extreme, just try to figure out what’s really necessary. Zebra scanners or field service add-ons are typical misses in the out-of-the-box set up I’ve seen needed for basic business operations.

But with a strong enforcer who can say no to things, who can push the team to maintain part count accuracy, installing an ERP doesn’t have to completely disrupt everything when things inevitably get messy. You’ll have a much easier time cleaning up messes when you limited the operational space from the get-go.

So keep it simple, make this thing walk before you try to make it fly and you can avoid it crashing, killing the morale of the entire ops department and blowing through all of next year’s software budget this quarter.

If you enjoyed this, SecretCFO has written a great thread on X for more tips on setting up an ERP, definitely worth subscribing to his X account and newsletter for great financial insight into managing a business.

Keep Reading

No posts found