The Automation Checklist Every NE Indiana Manufacturer Should Run This Year

Most automation advice is written for the Fortune 500. A thousand employees. A real IT department. A consulting firm on retainer.

If you run a 40-person shop in Wabash or Huntington, that advice is worse than useless. It is expensive and not useful, which is the worst combination there is.

I am writing this from Wabash. The version below is what I actually walk through with manufacturers in the corridor that runs through Wabash, Grant, Howard, Huntington, Miami, Wells, Whitley, Noble, and Allen counties. You can do the whole thing in an afternoon. You do not need to replace your ERP. You do not need a binder.

Start by watching

Before you write anything down, walk. Office first, then the floor. Forty-five minutes. No phone. No clipboard yet.

You are looking for three kinds of work:

1. Things one person does that nobody else knows how to do
2. Things that get done late or wrong often enough that customers feel it
3. Things that pull your highest-paid people off the work only they can do

That is your starting list. In a typical 50-person shop, I find five or six of these in the first hour. Three are usually worth automating. The other two or three are management problems pretending to be automation problems, and I will get to those.

Your ERP can already do more than you think

I have lost count of how many shops I have walked into where the ERP came with the building. Sage. Epicor. Global Shop. JobBOSS. Plex. Microsoft Dynamics. Whatever you have on the back office machines, you are using maybe 40 percent of it.

Call your vendor this week. Ask them, plainly, what features you are licensed for that you are not turning on. Then ask what they have added in the last two release cycles.

You will be embarrassed at the answer. That is fine. Some of the biggest wins I see for manufacturers in this region are not new software. They are turning on the module they have paid for since 2019.

This is not glamorous work. It is also free. Do it before you spend a dollar on anything new.

Find the data living in Bob’s head

Every manufacturing shop I have ever seen stores its data in three places. The ERP. A spreadsheet on someone’s desktop. And Bob.

Bob has been there 23 years. Bob knows which heat treat to call when the regular one is backed up. Bob knows the customer at the OEM in Auburn who actually approves the PO, not the buyer on the email chain. Bob is going to retire in four years, and when he does, you are going to have a very expensive problem.

For each of your top bottlenecks, ask one question: where does the information live? If the answer is “Bob knows,” that is not an automation problem yet. That is an institutional knowledge problem. You fix that first, by getting it out of Bob’s head and into something that survives him. Then you automate.

People skip this step and then wonder why their shiny new system is full of bad decisions. The system is only as good as what you fed it. Our assessment is largely about finding the Bobs before they retire.

Documentation is the new bottleneck

If you touch ag, automotive, aerospace, defense, food, or anything medical-adjacent, this is the section that pays for the rest of the post.

The documentation requirements are not getting lighter. PPAP packets. Certificates of conformance. Batch records. Lot traceability. AS9100 audits. Whatever your alphabet soup is, your team is doing more of it by hand than they should, and the customer is asking for more of it every year.

The rule I use with clients: if you cannot produce the document inside five minutes when someone walks in asking, that document is an automation candidate. Not “we can find it.” Produced. In five minutes. With the right signatures and revision.

That is the bar. It is not theoretical. The next time a Tier 1 auditor shows up, that is the bar they will measure you against.

Decide what stays human

This is the step nobody writes about because it does not sell software.

Some things should not be automated. Ever. The conversation with a 30-year customer when their order is going to be late. The talk with a long-tenured operator about a quality miss. The judgment call on whether to scrap a lot or rework it. The first phone call with a prospect.

Write down what you will not automate. Tell your team. Put it on the wall if you have to. It clarifies everything else, because now your operators know that the new system is not coming for the part of the job they actually care about.

I have watched shops lose good people because they automated the wrong thing. Do not be that shop.

What you do with the list

You will end up with a list. Maybe five items. Maybe ten. Most of them will not be automation projects in the way the trade magazines describe automation. They will be ERP modules to turn on, spreadsheets to retire, knowledge to extract from Bob, and audit trails to clean up.

That is the work. It is not glamorous. It compounds.

If you want someone to walk through that list with you against the seven operational areas I use with every manufacturer in the corridor, that is what our assessment is. Two to four weeks. Real numbers come out the other side, not a deck.

If you would rather pull up a system that handles the audit trail, the documentation, and the institutional knowledge piece in one place, look at the Desk. It is what I would have built for myself if I were running a shop in Huntington.

Or just book a call. I am in Wabash. I do not bill for the conversation, and I am happy to tell you whether what you found on your walk is worth chasing.