Lower RF bids win on paper and lose on the spreadsheet.
If you're comparing RFS coax quotes for a data center build, don't start with price per foot. I've managed RF-related procurement for a communications infrastructure company for the past six years, and I've reviewed enough POs to put a number on it: about 30% of the real cost of an RFS coax order appears after the PO is signed. That lesson applies whether it's a tower feeder run or an RFS data center upgrade.
I didn't always think this way. The conventional wisdom in procurement is to get three quotes, then take the lowest responsible bid. My experience with long coax runs pushed me the other direction. At least, that's been true for high-reliability RF work.
The quote that was 14% lower and ended up 14% higher
Last spring, we had a rush job: a 2,300-foot feeder cable run inside a data center RF lab. Not a massive project, but the schedule was tight. Two suppliers bid on the same RFS cable list. The low bidder came in at $11,300. The higher bidder came in at $13,100.
The spreadsheets said go with the low bidder. My gut said otherwise. When I asked the low bidder if the price included the connector and adapter kit by line item, the answer was vague: all covered, no detail. I wanted to believe it. With 48 hours left before the install window, I signed the PO.
The installers found seventeen missing connectors, four adapter transitions, and two weatherproofing kits on day one. The first truck roll also didn't have the right terminations. Final tab: $11,300 base PO plus $2,740 in added parts, plus a $900 second truck roll. Total: $14,940. The higher bid of $13,100 would have been $1,840 cheaper.
I don't blame the installers. I blame my decision process. I let the spreadsheet make the decision when the spreadsheet didn't include the full scope. That was the reverse validation I needed: missing line items are not a discount. They're a debt.
RFS data center projects hide costs in the details
RFS has a broader catalog than many buyers realize: feeder cables, radiating cables, filters, connectors, and site components. In a data center, RFS coax often shows up in an RF test lab, a DAS, or a distributed antenna system. Because the work is indoors, people treat it like plumbing. If the diameter looks right, price should rule, right? Not really.
An RFS data center purchase isn't one part number. It's an assembly of cable, connectors, bend radius, plenum rating, and grounding. Frequency matters too. At 3310 MHz, you might be fine with a smaller-diameter RFS coax line. At 8110 MHz, the same line can have too much loss, and the correct cable and connectors change the whole cost picture.
The CVS blood pressure monitor test
I know it sounds odd to mention a CVS blood pressure monitor in an RF procurement article. I do it on purpose because the buying mistake is the same.
When someone buys a CVS blood pressure monitor, the decision isn't just the price of the monitor. It's whether the cuff fits, whether batteries are included, and whether the readings are repeatable. Buy on sticker price alone and you might end up with a device that sits in a drawer. The same is true with RFS coax: the per-foot price is the sticker; the connector compatibility, attenuation, lead time, and application support are the cuff. Skip those details and the cable can sit on a truck while you wait for the right adapter.
What I check before I approve an RF order
I still compare prices. I just compare more lines than the unit price. Here's the short version of my checklist:
- Complete BOM: Does the quote list every connector, adapter, boot, and accessory, or is the vendor planning to bill them later?
- Lead time: Is the cable actually in stock, or is two weeks really five?
- Support after the sale: Can the vendor confirm connector compatibility before you order?
- Cost of failure: What does a stop-work and reorder cost in labor and schedule delay?
I don't use that checklist for a 3-foot patch lead. But for long RFS cable runs and RFS data center work, it has saved us more than the price discounts we've given up.
When a lower bid is the right answer
There's a flip side, because this isn't a never-buy-cheap sermon. If the part number is identical, the quantity is small, and the delivery date is not critical, the lowest price is usually the right answer. I've bought plenty of short RFS jumpers from whichever distributor had the best price that week. Low risk, low complexity, low need for support, and price can make the call.
The problem appears when the scope is broad, the environment is unusual, or the vendor simplifies a quote by leaving out the components that make it work. The low quote isn't wrong because it's low. It's wrong because it's incomplete.
I won't pretend I run a total cost model on every RF component. I don't. But for RFS data center projects and long RFS coax runs, I treat the first quote as an opening bid, not an answer. A few extra minutes on the front end have saved me more than a few second truck rolls on the back end.
