Hitting the PIM Ceiling: What Distributors and Manufacturers Can Do About It
Why your product data strategy stops mattering the moment you outgrow it
Jay Roxe from Inriver has a dare he gives everyone. Go home tonight, snap one of the little clips that holds a shelf in your fridge, and try to order the replacement online.
He’s done it. It’s miserable.
Now picture that same misery multiplied across a supplier’s entire catalog. Not a shelf clip. Everything. Your HVAC unit. Your industrial pump. Your electrical connector. Each one a search that ends in dead ends and frustration. That’s the moment you realize you’ve hit the PIM ceiling.
I spend a lot of time talking to manufacturers and distributors, and when they talk about product data problems they usually say the same thing; we’ve hit a ceiling. The business is running fine. Growth is happening. But the tools that got you here can’t take you further. Everything’s held together with workarounds. Everyone is still living inside spreadsheets. Someone knows which spreadsheet is the source of truth, but nobody else does. And if that person wins the lottery and moves to the Canary Islands, you’re in real trouble. Love that line.
That’s not a cute problem anymore. That’s a revenue problem.
The Ground Shifted and Nobody Noticed
Here’s what’s changed since just before COVID. The number of channels a manufacturer sells through has more than doubled. So has the number of attributes you have to track per product. More places to be, more to say in each one, less time to get there.
You’ve got Amazon. You’ve got your own website. You’ve got distributor portals. You’ve got marketplace partners. You’ve got answer engines.
Yes. Answer engines. That’s the part that caught fire this year.
More buyers are building their shortlists from whatever Claude or ChatGPT or Gemini tells them. And here’s the thing those engines don’t do; they don’t read your marketing copy. They don’t skim your website. They read structured, materially correct product data. If your data says a part weighs five pounds and it actually weighs fifty, an answer engine tells your buyer it weighs five pounds. They order based on that. The wrong part arrives. The machine sits idle. Your customer calls someone else next time.
That’s the moment a team realizes Excel, “the second-best tool for every job ever invented”, isn’t cutting it here.
Excel worked fine when you had one hundred SKUs and two sales channels. It doesn’t work when you’ve got fifty thousand SKUs, each one needs to live in seven different places with different attributes, and a machine is reading the data to make buying decisions.
The Revenue Leak Nobody Can See
For manufacturers, there’s a revenue stream that’s easy to forget about. Aftermarket parts and service.
You break a cheap pair of shoes, you just replace them. Nobody scraps a multimillion-dollar HVAC unit sitting on a rooftop. They don’t throw away a combine sitting in a field because one part broke. They need the exact right replacement, accurately described, or the whole machine sits idle. They need it fast. And they need to be able to find it.
Scale the broken-fridge-clip problem up to that, and you can see both the pain and the money leaking out of it. A large percentage of manufacturers make as much money on aftermarket parts and service as they do on the original equipment. If your product data is buried or inaccurate, you’re handing that revenue straight to the gray-market seller who’s happy to sell the right part to your customer instead.
The teams handling this well treat it as a layered thing. What’s your underlying data foundation, where’s it coming from, and can you trust it? Are you wired to your ERP and your downstream merchandising in a structured, automated way? Have you built the right enrichment and workflows on top? That’s the difference between a system of record and a system that actually works.
The Maturity Gap Between Spreadsheets and Legacy Systems
Here’s what’s interesting about the teams that are stuck. The difference between the spreadsheet crowd and the legacy-PIM crowd is mostly maturity, because the underlying problem is the same.
At least the legacy-system folks understood that product data should live in one golden record. They built that. The spreadsheet world is scattered and hand-cranked; the data lives in a dozen tabs and nobody knows which one’s current.
But even a consolidated catalog can be consolidated inside one person’s skull. You know the type. The person who knows which version of the spreadsheet is the source of truth. The one who has every workflow mapped in their head. The one nobody else can fully replace.
If that person leaves, the business is in trouble. If they get sick, if they move, if they win that lottery; the entire operation suddenly runs on institutional knowledge instead of systems. There’s no contingency plan here, just vulnerability.
The numbers make the case that you can’t ignore it. In a study of about sixty-five customers, Inriver found products going to market roughly thirty percent faster after switching from legacy systems to a modern PIM. That’s transformative.
There’s a question Jay always asks; “of the SKUs that should be ready to sell right now, what percentage actually are?” About a third of companies on older systems have less than eighty-five percent of their catalog publish-ready. Think about that. One in three manufacturers can’t actually sell a third of their inventory because their product data isn’t complete or accurate enough to publish.
That’s a lot of orders handed straight to someone else.
What Actually Matters Now
The shift in what a PIM even is, is real. It’s growing up from a system of record into a system of work. Modern setups are being driven by agents and workflow orchestration, not just data storage and retrieval. Legacy teams, whether they’re on a crusty old PIM or a wall of spreadsheets, they’re the frog in the slowly heating pot. The pain built up so gradually they never stopped to ask how the workflow should actually run.
The answer-engine moment changes the calculus. When machines are reading your data to help buyers make decisions, product data quality becomes a go-to-market function. It’s not a back-office thing anymore. It’s not something IT owns and everyone else ignores.
It’s marketing. It’s merchandising. It’s revenue.
Two Exercises That Show You What’s Actually Happening
For the VP of product, merchandising, or eCommerce who suspects money’s leaking but can’t see where, Jay offers two exercises.
First, staple yourself to an order. Follow one product through every single step of the process. Watch where it comes from. Watch how it gets enriched. Watch where it sits waiting. Watch it go out to each channel. Watch the gap open up between how it’s supposed to work and the workarounds holding it together. You’ll see duplication. You’ll see rework. You’ll see time leaking out.
Second, run a quick answer-engine assessment. Search for some of your core products in Claude, ChatGPT, Gemini. See how you actually show up. See what information the AI gives your customer. See where it gets it wrong. See what’s missing. That’s your customer’s buying experience right now, and if it’s bad, you’re not competing against another manufacturer. You’re competing against an answer engine that confidently told them the wrong thing.
Both exercises will likely show you the same thing; the business has outgrown the tools.
The Best Time to Plant This Tree
Jay says the best time to plant this tree was five years ago, when the channels were fewer and answer engines didn’t exist. The second-best time is today.
Because the problem doesn’t get easier. More channels are coming. More attributes per product are coming. Answer engines are getting smarter about reading product data. The companies that wait for the pain to become unbearable will be fixing this in crisis mode, which is expensive and slower than fixing it now.
The teams that are thinking about this today, building a single golden record, connecting their ERP to their merchandising, making sure their product data is accurate before it ships, they’re going to outrun the ones still living inside spreadsheets.
Listen to Episode 9 of Data vs. Commerce to hear Jay Roxe from Inriver go deeper on this. How modern PIM actually works. Why the aftermarket parts business is where a lot of manufacturers are leaving money on the table. What product readiness actually tells you about your operation. And what those two exercises reveal about the gap between your process and your tools.
Because right now, somewhere in your catalog, there’s a fridge clip nobody can find. And if you can’t find it, neither can your customer. And if your customer can’t find it, they’ll find someone else.
Data vs. Commerce is the podcast where we talk about the real stuff happening in distribution and manufacturing. Where the strategy actually meets the operation. Where the people who run these businesses share what they’re learning, what’s breaking, and what’s working.
New episodes drop weekly. Subscribe wherever you listen.
