Three years ago, I approved a quote for 47 Siemens S7-1200 PLCs from a distributor we'd never used before. It was 18% cheaper than our usual supplier. I felt like I'd done my job.
Fourteen months later, I was sitting across from our operations director explaining why we'd spent an extra $31,000 on that "savings."
I'm a procurement manager at a system integration company—200 people, roughly $480,000 a year on automation hardware. I've negotiated with more than 30 vendors and tracked every order in our cost system for the past seven years. The biggest lesson I've learned is one nobody taught me at the start: the unit price on a PLC quote is the smallest part of what you'll actually pay.
The Problem You Think You Have
If you're like I was, PLC procurement looks about like this. Project comes in. You get 2-3 quotes. You compare per-unit prices. You pick the lowest. You file the PO. You move on.
Same part number, same spec sheet, so lower price must be better. Right? It's the most natural assumption in the world.
It's also wrong—not in the vague "quality matters" way people usually mean, but in a measurable, budget-tanking way.
What the Quote Actually Leaves Out
Here's what that 18%-cheaper quote excluded, either explicitly or simply in practice:
- Software licensing. TIA Portal seats through the new distributor came with a different bundle structure. The step where the Siemens licenses were actually transferable to our existing project files? Not so much. Add $4,100 in re-licensing within the first six months.
- Firmware mismatches. The batch of S7-1200s came with older firmware than the G2 catalog implied. Our techs didn't catch it until commissioning. Three weekends of extra engineering time—roughly $6,800.
- Support calls that weren't included. Our usual distributor bundles about 20 hours of technical support into the first year. The new one didn't, and their support line routed to a regional office that didn't know our project. Every standard question became a paid ticket: $220/hr.
- Spares sourcing. Two units failed at the customer site. Replacements had a 3-week lead time because the distributor didn't hold local stock. We sourced replacements from a second distributor at 40% premium to avoid missing a contractual deadline.
Total extra spend on that "cheaper" order: roughly $31,000. Original savings: about $9,500. The math isn't subtle.
The Deeper Problem: Unit-Price Optimization
I used to think the mistake was buying from an unfamiliar vendor. That wasn't it.
The actual mistake was optimizing individual transactions instead of the portfolio. Every order looked like a smart deal in isolation. Together, they created a maintenance and support environment that was 2-3x more expensive than it needed to be.
This pattern is easy to miss because it's invisible in the moment. Here's how it actually plays out:
- You buy the cheapest quote for Project A. Fine—it ships, it works.
- Project B needs Beckhoff for a specific OEM application. You buy Beckhoff because the integration is cleaner. Now your maintenance team needs fluency in two ecosystems.
- Project C needs an S7-1500 for a process line. You buy from a new distributor because they quoted low. New support path, new lead times, new firmware quirks.
- Project D needs spares. Now you're keeping spares for three brands, three firmware versions, three support relationships.
None of these decisions feels wrong at the time. All of them are individually defensible. But by year three, our maintenance team was spending maybe 20% of their time on tooling confusion and vendor logistics—time that doesn't show up on any quote.
This is where the Beckhoff-versus-Siemens discussion usually comes up in our project kickoffs. Here's what I've learned after using both. Neither is the "right" answer, and I mean that. Beckhoff is genuinely strong at certain motion-control and high-speed I/O applications where its EtherCAT ecosystem is cleaner. Siemens is strong at process automation, distributed control, and—this part matters for cost—line-replaceable-unit availability across a huge installed base.
If you're running 90% Siemens and someone wants Beckhoff for one machine, that's a real cost, not a "best of breed" win. If you're starting fresh with no ecosystem at all, Beckhoff deserves a serious look. The honest answer depends on where you already are.
"Track the three-year cost of a decision, not the line item. The line item has already been optimized by someone else's sales team." — my operations director, during that conversation I'd rather not repeat
What This Actually Costs Over 5 Years
I went back through our 2021-2024 hardware POs to quantify this. For our $480K annual automation budget, here's what I found:
- Unit price variance between distributors: $18,000-$24,000/year in potential savings if everything else were equal.
- Cost of vendor fragmentation (training, spares, support inefficiency): estimated $110,000-$140,000/year in additional labor and premium sourcing.
- Downtime from incorrect firmware/config: 6 incidents in 3 years, each averaging 14 hours of production-equivalent loss.
In other words, the savings I thought I was chasing were maybe a fifth of the cost I was creating. The "cheap" option wasn't cheap. It was expensive in a different column.
And the pattern shows up everywhere, not just in PLCs. Our maintenance team's small tooling purchases have the same problem. We buy $35 multimeters, and technicians can't reliably identify the continuity symbol—that little diode-with-arc-lines icon—on the cheap LCD, so troubleshooting takes longer. We buy $18 gel battery chargers for the 12V backup systems on our telemetry cabinets, and they fail inside a year. Someone buys the cheapest cigarette lighter battery charger for the field vehicle kit, it dies in cold weather, and the whole vehicle is short one charger for a month.
None of those are big purchases. Together maybe $4,000 a year. But the labor cost of working around them is 5-10x that.
What I Do Now
I changed three things, and none of them required a corporate initiative:
1. I don't compare quotes on unit price anymore. I compare on a 3-year TCO sheet that includes licensing, support, estimated downtime, and spares lead time. The spreadsheet took two days to build. It's saved far more than that.
2. I've made ecosystem consistency a soft requirement. If a project can be delivered on our existing Siemens stack—which for us means S7-1200 G2 for most mid-range machines and S7-1500 for process lines—I'll pay up to 8% more for that consistency. Above 8%, the fragmentation cost stops being worth it.
3. I buy from two distributors, not five. Competition between two known vendors keeps prices honest without spreading our support and stock thin.
The S7-1200 G2 specifically has earned its place in our standard build. The protection features and tighter TIA Portal integration mean less commissioning time and fewer support calls. But I want to be upfront: it's not the right answer for every shop.
If most of your automation is motion-control-heavy with high-speed synchronization, Beckhoff's EtherCAT ecosystem is worth a serious look. If you're supporting a lot of legacy S7-300 installations and can't justify a migration, staying on proven hardware—even if it's not the newest catalog—is often the cheaper call. And if you're under 15 PLCs total, the ecosystem argument mostly disappears, and per-unit price matters a lot more. The honest answer depends on where you already are.
What I've stopped pretending is that any single column on a quote tells the whole story. It never does. The vendors know this. I just wish I'd learned it before I cost my company $31K to find out.
