Kicker Search Volume Lied to Us: How We Rebuilt Our Parts Catalog Around Buying Intent
The Short Answer
We stopped using raw search volume to decide what to stock. That one change cut our dead inventory from 12.4% to 4.1% in about 18 months, and kept roughly $47,300 in goods out of the write-off column in 2023 alone.
Search volume tells you what people are curious about. Purchase history tells you what they'll buy. On a spreadsheet those two things look alike. In a warehouse they behave nothing alike.
If you only have time for one line: before you stock a SKU because a keyword looks hot, pull the actual queries that convert on your own site. The gap between “searched” and “bought” is usually ugly, and it's where most of our bad orders came from.
Why You Should Believe Me (And Where I'm Limited)
I run procurement for a 45-person aftermarket parts distributor. I've managed our purchasing budget — about $2.1M a year — for eight years. Our catalog covers car audio, exhaust systems, underbody protection, and installation tooling. I've signed off on somewhere north of 1,900 POs since 2017.
I'm not a data scientist, so I can't speak to intent-classification models or anything that needs a stats degree. What I can tell you from a buyer's seat is what happened to our numbers when we changed the filter.
Everything I'd read about assortment planning said to follow demand signals, and search volume was the easiest one to grab. In practice, for a mixed catalog like ours, it was the worst one. Not because the data was wrong — because it was answering a different question than the one I thought I was asking.
One more thing before the case studies: our dead inventory number was never just the cost of the goods. When I audited our 2023 write-offs, $47,300 was the goods number. Add inbound freight, the shelf space, and the eventual markdown, and the real hit was closer to $61,000. That's the number I bring to leadership now, because it's the honest one.
The Kicker Trap: A Brand Name That Isn't a Buying Signal
Here's the one that embarrassed us.
Kicker is one of our biggest lines — subwoofers, amps, marine audio, the whole catalog. So when our keyword platform showed “broncos kicker history” and “nevis rams kicker” pulling over 100,000 combined monthly US searches — 119,000 when I pulled it in March 2024 — someone on our team flagged it as a demand spike worth acting on.
It wasn't. Those are NFL queries. Broncos kicker history is a football topic. Nevis rams kicker — same thing, different team. Nobody typing those phrases is shopping for a 12-inch subwoofer. But the word “kicker” is in there, the volume is huge, and if your reporting only joins on keyword text, you get a false positive that looks identical to a real one.
We committed $6,800 of slow-moving Kicker inventory off that signal in Q1 2024. Took eleven months to clear. About $900 of it went out the door at cost.
The question everyone asks is “what's the search volume?” The question they should ask is “what's the intent behind it?”
The fix was dumb and cheap. Every keyword signal now has to be paired with a commercial modifier — buy, price, install, wiring, replacement, fitment, review. If the phrase doesn't carry at least one, it goes in a research bucket instead of a purchase bucket. “Broncos kicker history” fails that check in about two seconds. “Kicker subwoofer 12 inch marine” passes. Same brand, same first word, totally different decision.
Most distributors focus on keyword volume and completely miss intent classification. It's the same mistake as judging a vendor by unit price and ignoring everything that happens after delivery.
Model-Year Timing: 2016 Tahoe Exhaust vs. 2026 Tacoma Skid Plate
Two products, two directions, same underlying lesson.
2016 Tahoe exhaust system
We picked up a 2016 Tahoe exhaust system line in 2019 because the vehicle was everywhere and the OEM systems were rusting out fast in the Northeast. Good call early. Bad call by 2023.
What we missed is that aftermarket exhaust demand tracks vehicle age in a curve, not a flat line. Owners of a nine-year-old Tahoe either keep it stock or they go fully custom. The bolt-on cat-back middle ground dries up first. Our turns on that line went from 62 days in 2021 to 94 days in 2024. We still carry it — it moves, slowly — but at roughly 40% of the depth we used to hold.
There's a cost angle here that gets skipped: returns. Every return carries outbound shipping, return shipping, inspection, and then either restocking or a markdown. According to USPS published rates effective January 2025, even the cheapest First-Class Mail large envelope is $1.50 to send, and that's the floor — real parts move Priority or Ground at multiples of it (source: usps.com/stamps). Stale categories produce more returns because buyers guess wrong more often. It compounds.
2026 Tacoma skid plate
The opposite mistake is buying in too early. In February 2025 a supplier offered us a pre-order on a 2026 Tacoma skid plate mold — $4,200 for the tooling, plus a 12-month minimum commitment.
We said no. The reason was simple: we couldn't test-fit a part for a truck that wasn't in our bay yet. Mounting points, oil drain access, sway bar clearance — these things get revised between the press photos and the production line. A skid plate that doesn't clear the drain plug is a return, not a sale.
We took the slot three months later instead, after a customer rolled in with a delivered truck. Cost us a small premium. Saved us a $4,200 mold we might have had to redesign twice.
That's the prevention-over-cure math, and it's boring: five minutes of fitment verification beats five weeks of returns processing. Every time.
Who Owns Power Tool Brands (And Why It Broke Our Vendor List)
Different problem, same category of error.
We buy installation tooling — impact drivers, torque wrenches, heat guns. Our old vendor list treated every brand as a separate supplier. Then a rep told us something useful: a lot of the “competing” brands on our list share a parent company.
Rough ownership map, as of Q1 2025:
- TTI owns Milwaukee and Ryobi
- Stanley Black & Decker owns DeWalt, Craftsman, and Porter-Cable
- Chervon owns EGO and Skil
- Makita is independent
- Bosch owns Bosch and Dremel
Why it matters for cost control: if two of your “vendors” are the same company, your leverage is smaller than your vendor count suggests. We thought we had nine tool suppliers. After mapping ownership, we had six. That's a different negotiating position when it's time to re-quote a contract.
It also matters for parts commonality. Two brands under one roof sometimes share batteries, chargers, or accessories, which changes what you actually have to stock. I should add that this cut both ways for us — consolidation freed up about $3,100 of shelf inventory we didn't need to duplicate.
Brand ownership isn't printed on the box, and it shifts with acquisitions. If you build this map yourself, check the parent company's investor relations page and date-stamp every entry. Ours says “as of Q1 2025” at the top, because I've been burned by undated lists before.
One more angle. The FTC's business guidance on advertising requires claims to be truthful, substantiated, and not misleading (ftc.gov/business-guidance/advertising-marketing). That applies to how a supplier describes its own brand relationships too. If a rep is vague about who owns what, treat the vagueness as information — it usually is.
Where This Approach Doesn't Work
Two honest limits.
First, search volume isn't useless. For commodity parts with intent baked into the query — “buy 2016 Tahoe exhaust,” “12 inch subwoofer price” — volume and demand move together closely. We still use it there. The filter I described is for mixed or ambiguous terms, the kind where a single word like “kicker” can mean three different things to three different searchers.
Second, this only works if you have conversion data. If you're a newer distributor without 12+ months of order history, you can't build the intent filter yet. My suggestion in that case is to start narrow: buy the smallest order quantity your vendor will allow, track returns and turns for 90 days, and don't commit to tooling or minimums until you have your own numbers. The $4,200 mold we didn't buy and the $6,800 in Kicker inventory we shouldn't have bought are the same lesson from opposite ends. The check is cheap. The correction isn't.