A retaining ring worth a few cents travels a remarkably long road through many manufacturing companies. It starts as a line in the Inventor BOM, gets retyped into an Excel sheet for costing, is looked up in the ERP system for its item number and is confirmed to purchasing by email. Four programs, four copies of the same information, and at every handover a chance that something changes along the way that should not change.
That constant context switching between engineering software has become so normal in many design offices that nobody notices it anymore. Yet every switch between Inventor, Excel, ERP and email is really a manual interface between systems, operated by someone who was hired to do very different work. In this article you map those switches and replace them one by one with data that flows straight out of Inventor: mostly through agreements and the software you already have, and where it fits with a toolbox like Thundercad.
Every switch is a manual interface
Anyone who leaves Inventor to look up an item number in the ERP system is doing exactly what a system integration does: pulling data out of one package to use it in another. Except that here the connection runs through eyes, short-term memory and a keyboard. That works, until it does not: an 8 becomes a 3, two digits swap places, a line gets skipped because the phone rang.
The nasty thing about these retyping errors is how late they surface. A wrong item number looks fine in Excel and fine in the ERP; it becomes visible when a box with the wrong retaining rings shows up at assembly. By then the error can no longer be traced back to the moment it was made, and the familiar hunt begins: was it the model, the list or the ordering system?
Run the numbers for an average assembly. Assume a BOM of a hundred and twenty lines, sixty of which have to go into the ERP, at roughly twenty seconds of looking up and entering per line. That is twenty minutes of pure data transport per assembly, and nothing has gone wrong yet; a single swapped digit in those sixty lines easily costs a multiple of those twenty minutes further down the process.
First map where you switch
Improvement starts with seeing where the switches are. In most design offices they fall into four fixed categories:
- Copying BOM data: retyping or pasting lines from Inventor into Excel for costing, purchasing or work preparation.
- Looking up items in the ERP: does this part already exist, what is its number, what is the description, who supplies it.
- Hunting for files in Explorer: finding the right project folder, retrieving yesterday's export, checking whether a PDF of revision B already exists.
- Emailing data around: a colleague asks for a weight, the customer wants a parts list, purchasing wants written confirmation of the material.
Walk through a normal working week and note per category how often it occurs and, more importantly, which data gets retyped. No need for stopwatch precision; the proportions are enough. Almost always the same picture appears: the BOM route and the item lookups are the two big flows, the rest is small change that adds up.
The BOM: from retyping to exporting
The BOM is almost always the biggest data flow between Inventor and the rest of the company, and therefore the place where cutting manual transfer pays off most. As long as that list leaves the model through copy and paste, or worse, through retyping from a second screen, every costing sheet and every order rests on a handmade copy of the model.
The solution is to let the BOM come straight out of the model, in the shape the recipient expects. With Export BOM from Thundercad you export the bill of materials of an assembly to Excel in one action, in your own template, so costing and purchasing always receive the same file in the same layout. Which columns belong in it and how to set up such a template is covered in BOMs from Inventor to Excel: always in your own template.
Even more important than the minutes saved: the export does not lie. What is in the model is in the list. When the assembly changes, you simply run the export again, instead of repairing an Excel file nobody dares to vouch for anymore.
Curious how much data transport is hiding in your BOM route? Export your next bill of materials straight into your own Excel template with Thundercad and compare it with the old routine.
Try 30 days freeConsulting items without leaving Inventor
The second big flow runs the other way: from the ERP to the engineer. Does an item already exist for this tube? What is the number of that motor, and is it still current? In many companies every such question means: minimise Inventor, open the ERP, search, jot the number on a scrap of paper, back to Inventor, type it in. Ten times a day is not unusual.
You reverse that route by bringing the ERP into Inventor instead of the other way around. With Article Manager you consult the items in your ERP system directly from Inventor and place the found item straight into your assembly, with the right number and description attached from the start. The scrap of paper disappears, and with it the typo that was waiting to happen.
A setup like this stands or falls with the quality of the item data: one item, one number, one truth. How to organise that, and how to prevent CAD and ERP from each keeping their own version of an item, is something we covered earlier in Article management between CAD and ERP: one item, one truth.
The small switches: Explorer and email
That leaves the small switches. Individually they are trivial; together they fill a surprising share of the day. Hunting for files in Explorer is a classic. The remedy has two parts: a fixed, predictable place for every export, so searching becomes looking up, and as few clicks as possible to get there. Agree per project where exports land, and stick to that even when things get busy. Whoever wants to jump from the open document straight into its folder can cover that with Go To Folder; it saves only a handful of clicks each time, which is exactly why it adds up.
Email, finally, is usually not a switch of its own but a symptom of the previous three. Whoever asks for a weight, a parts list or a confirmation of the material is really asking for data that was not findable or not trustworthy. As BOMs start coming from the model and items have a single truth, that mail stream dries up by itself. The question "which version is this?" does not go away by answering faster, but by making it unnecessary.
Frequently asked questions
Is switching between programs not simply part of the job?
Partly, yes. You will always work in multiple systems, and that is fine as long as each system does what it is good at. The distinction is between switching to decide and switching to transport: the first is work, the second is retyping, and that second part is what you can eliminate.
Which data flow should I tackle first?
Pick the flow with the most retyping and the most expensive errors; in practice that is almost always the BOM. It touches costing, purchasing and work preparation at the same time, so one improvement pays off three times over. After that, item lookup is the logical second step.
Is this worth it if our ERP has no link to Inventor yet?
Yes. A bill of materials that lands in a fixed Excel template through Export BOM is faster and less error-prone than retyping even without any integration, and almost every ERP system can import such a file. If you want to see what that looks like on your own projects, you can try Thundercad free for 30 days; through the Dashboard you can then connect it to your ERP whenever you are ready.