Have You Hit the PIM Ceiling? | Ep. 9

Episode - Have you hit the PIM ceiling? - Transcript

Matt: Welcome to Data Versus Commerce, where we explore the messy middle between database and doorstep. I’m Matt Johnson.

Floyd: And I’m Floyd Blaikie. Let’s dig in. Okay, we are here with Jay Roxe. It’s the first time I’ve said it out loud — it sounds like a diagnosis instead of a name. Did I get that right, Jay? You want to tell us a bit about yourself?

Jay: Floyd, great to be here with you and Matt. As somebody said, it’s not just a name, it’s a sentence.

Matt: With an exclamation point on the end, Jay — with an exclamation point on the end. I’ll just point that out.

Floyd: Yeah, Jay rocks.

Jay: Don’t I wish?

Floyd: I’m sure you do, I’m sure you do. So tell us about yourself, Jay — why are you here, what do you do, and what are you interested in?

Jay: Floyd, Matt, everybody listening — name’s Jay Roxe. I work at inriver on the marketing team, and I’m really just interested in the shifts we’re seeing in the market — the move to people trying to bring products to market faster, what’s required, and how the commerce landscape is changing.

Floyd: Very cool — I’m also interested in that too.

Matt: Interested in that too, Jay.

Floyd: It absolutely is, and that’s why we’re having you on here — we’re kindred spirits. I’m also in a marketing role in my day job, and one of the exercises I do on an ongoing basis is listening in on our own customer calls and some of our sales calls, and really getting a feel for how prospects and customers talk about their problems.

I think that’s so important, and one of the phrases I keep hearing, especially in the last six months, is this feeling of hitting a ceiling. They’re chugging along, but they feel like they’ve outgrown their tools or their processes. Everything is run by workarounds, everyone’s living inside an Excel spreadsheet — which I think is how most businesses operate. So I want to dig into that a bit with you, given your expertise in the PIM space, and talk about what you see when folks come to you wanting to know more about moving to a modern PIM. What’s the biggest trigger you see when you’re talking to manufacturers, distributors, retailers, and it’s keeping them up at night — “I might need to think about this PIM thing.” What’s going on there?

Jay: I think there’s a couple of different things, and let’s acknowledge that for a lot of these folks, the landscape has changed dramatically. If you go back and look at the world pre-COVID, or even just coming out of COVID, what we’ve seen from our customers is the number of channels they’re going to market through has more than doubled. And as they’re doing that, the number of attributes they need to store on each product has also more than doubled. So it’s a much more complicated landscape, and at the same time, there’s huge pressure — for manufacturers, distributors, retailers, everybody — to take products to market more quickly and recognize that value a lot faster.

That was sort of the substrate that’s been building for a couple of years, and then it got turbocharged this year as AEO, GEO, and that kind of work became core to the buyer’s journey. More and more people are building their shortlists, looking for product information, and doing their selection process based on what Claude, ChatGPT, or Gemini tells them they should go after. And your marketing copy isn’t necessarily the thing those tools are consuming and building their lists on — they’re really consuming structured, materially correct product data. That’s, I think, some of what’s causing people to say, “Hey, wait — Excel is the second-best tool for any job ever, but maybe it’s not serving me well as we go through this.” Matt, I’ll toss it back to you — curious if that resonates with what you’re seeing too.

Matt: Yeah, it’s exactly right, Jay. Thinking specifically as a manufacturer, it’s a great example, because there are more channels than ever before. Couple that with the expectation around how to serve product information to those channels — keeping it consistent is as important as keeping it up to date, and both are as important as making sure it’s complete across those channels.

What I’ve been seeing is that manufacturers in particular are overrun with channel requests, and they don’t have a good process for renewing or refreshing those product data feeds. They don’t have the organizational structure to support 5, 10, 20-person teams working on merchandising spreadsheets. So there’s a real need — and even when manufacturers have a PIM, maybe a legacy solution they implemented seven or eight years ago, what they’re finding is it got really stale, and they stopped using it to its full capabilities. Some of these platforms haven’t evolved to think about how data needs to be structured for LLMs, for the modern buyer, and they’re stuck. We’ve seen it before, and we’re seeing it more and more today, to your point. So how do you get unstuck? Is it a platform switch? Is there something else going on? I’d be curious to hear what the best manufacturers are doing to actually take action on these problems.

Jay: I think part of the key question for manufacturers — and I’ll come back to your direct question in a second — is there’s a whole different revenue stream they need to be conscious of. Look, if you have a pair of shoes or something like that and it doesn’t work, you’re likely to just replace it. But if you have a multimillion-dollar HVAC system or a combine out in the field, you’re not going to chuck it out. You need to be able to accurately get the aftermarket parts and that service environment working — and it’s hard.

I have a challenge I give everybody to make this real: go into your refrigerator tonight, break one of those little clips that holds the shelf in place, and go try to order the replacement part. I’ve done this — it’s not easy. When you blow that out to a manufacturing level, we’re seeing a lot of manufacturers who’ve woken up to the fact — or have known for a while — that their PIM has to serve this entire environment of aftermarket parts and replacements.

I think the people doing the best on this are the ones who realize it’s really a layered project. What’s your underlying data foundation? Where are you getting it from, and can you trust it? Are you connected to your ERP instance? Are you connected to your downstream merchandising platforms in an automated, structured fashion? And then, have you built the right enrichment and the right workflows on top of that?

There’s a fundamental evolution happening in PIM, driven by agents and the other things we can now do with workflows — PIM is evolving from being a system of record into a system of work. When we talk to people on legacy systems, whether that’s a legacy PIM or legacy Excel, they’re frogs that have been boiled slowly, if you’re familiar with the analogy — the pain of the work has just built up, and they haven’t even noticed it until they stop and ask, “How should this be working? What should our workflow be?” That orchestration and governance piece is really where we see the best-in-class spending a lot of their time.

Floyd: First of all, I don’t have to break my own fridge — I’ve got kids who do it for me. I had this exact problem — I think we talked about it in the first episode of the podcast — where just trying to replace a part from my own dishwasher became an absolute nightmare, because the product information wasn’t up to date. They’d switched from a three-pin to a four-pin connector at some point in the model year, and it was an expensive, annoying mistake for me. That feels like a “getting to market accurately” problem. What’s the difference between what people with legacy PIMs are dealing with and what people running on the second-best tool for everything — Excel — are seeing? What’s the feedback they’re getting from their customers, their buyers? Is it the same — “we can’t find your stuff accurately,” “you can’t get to market fast enough,” “we can’t publish your catalog fast enough”? Or are there differences between legacy-PIM strugglers and Excel strugglers?

Matt: I’d say, Floyd, in answer to that, it really is a matter of digital maturity — how the company thinks about tackling these problems. At the end of the day, they’re the same problems. It comes down to whether we have the discipline to implement governance and process inside a system of record that, as Jay pointed out, has become much more than that. Now we can orchestrate complex workflows, now we can manage the entire product life cycle, whereas before, all of this usually lived inside somebody’s head, inside spreadsheets.

The difference between a company using spreadsheets and one using a legacy system is that at least the legacy-system companies understand the importance of consolidating product data into one golden record. The spreadsheet world is disconnected, highly manual. So step one: okay, you created a system, great — or you have a PIM, awesome. The question I’d ask, and Jay can speak to this: does that system actually fit the way product information needs to move through the digital supply chain? In other words, is there process, is there governance, is there automation? Automation, I think, is the game changer with modern PIMs like inriver.

Jay: Matt, 100% agree. What we find when we talk to customers still on Excel is that in many cases they have consolidated product information, but it’s consolidated in somebody’s head. There’s the one person who knows which spreadsheet it is, or which people to talk to for the information, and they have all the workflows mentally mapped. If that person wins the lottery and moves to Hawaii, there’s a massive challenge for the organization.

So understanding that workflow piece, building the business case — okay, if I actually invest here, first cut, it’s not only a reduction in work, but the possibility of reducing customer returns, reducing churn, launching faster. We did a study where we interviewed about 65 customers and found that, on average, they were launching products about 30% faster once they made the transition — that’s a meaningful difference.

When we’re talking with customers looking to make this transition, one of the questions I always ask is, “Of the SKUs that are publish-ready, that should be available to be sold, what percentage actually are?” We find about a third of organizations on older systems have less than 85% of their SKUs ready to sell — that’s a pretty meaningful gap.

The other thing organizations are starting to think about is the future. I think a lot of this moves toward a more prompt-based, MCP-driven environment, where you have agents executing across the various stages of your workflow and humans validating and governing. That’s not happening overnight — it’s a multi-year transition, particularly for larger organizations. But the starting point is now, and figuring out your strategy for getting there.

Matt: Yeah — the starting point was five years ago. I often think the exciting part that a lot of distributors and manufacturers don’t think about is the long-tail opportunity you get once you have an automated process for bringing products to market. You pointed out rightly, Jay — we, as an industry (and I include myself here, guilty in the past) display a fraction of what our customers actually need. So we’re stuck in a manual order-inquiry, RFP process, where I have to go to the rep to see if a part’s available. That’s the full dishwasher experience — you thought you could trust the website, turns out you probably should have just called in.

That’s the opportunity: how many dollars are we leaving on the table when we don’t bring our complete catalog to market? That’s where we need to think about the big-picture strategy — what’s our go-to-market strategy, how are we working with our channel partners — and then take that into a governance process and ultimately implement a system of record that lets us execute on that vision.

Jay: I’d say both system of record and system of work, to execute on that vision — because product data is more than ever a competitive asset. If you’re a manufacturer and you don’t have your complete product catalog online, someone offering gray-market parts is going to sell Floyd the replacement three-pin piece she needs for that dishwasher — and carry that forward into combines, HVAC units, all of it. It’s also, for everybody, about how fast you can take a new idea to market and reflect it accurately in every channel — that’s being magnified right now.

I’ll echo your point — the starting point was five years ago, but there’s a proverb that the best time to plant a tree was 20 years ago, and the second-best time is today. For organizations that haven’t started this transition, they’re not alone — there are many in that same boat, starting to figure out their strategy.

Matt: Love it.

Floyd: So if I’m a VP of merchandising, or a leader in channel sales, or a CTO — something I’m hearing a lot about is sellers missing out on sales because their products and product information aren’t ready for agentic or LLM search. But they don’t know that, right? Because you don’t get the bounce rate on your website anymore, you don’t get the digital tells that the sales aren’t happening. So how could I, in that kind of position, understand what’s being left on the table — the dollars being left on the table, to Matt’s point? How do I know that?

Jay: I have an exercise I love, and I’d encourage any merchandising manager to go through it: staple yourself to an order. Take something your team is looking to bring to market and follow absolutely every step of the process, and what the results are. I’ve always found it incredibly eye-opening — the difference between how the process should work and the workarounds that have developed to make it sort of work, which may be causing a huge amount of latent pain.

I’d also suggest people take advantage of some of the free tools available just to get a quick pulse check. We’re working with customers and interested parties on doing AEO assessments — free, quick, here’s how you actually show up in an answer engine. It’s a very insightful anchor point for understanding your current state in modern commerce. There’s a lot organizations may choose to do from there, but that understanding of the workflow, and of how you’re perceived in the market, are the key points.

Matt: I often think part of the advantage we have today is that the difficult work that used to be required to get people in the room, or to staple yourself to the order, to the workflow, can be so amplified with AI. A lot of times, people would come to us, and that’s where we’d start — we’d say, “Let’s map out the digital supply chain, let’s map out all the processes required to bring something to market, and let’s identify the points of friction.” Theoretically, you could do all of that within a few days — by ingesting data from your own organization about your operations, putting it into an LLM, mapping the process, and taking that to a system integrator like us and saying, “This is my process — let’s re-engineer it, and let’s start implementing the system of work that will help me eliminate all the manual friction so I can tackle the list of stuff we never get around to actually doing.”

It’s not about — and this is a great example — in merchandising, this is definitely not a case where AI is replacing anybody. This is a role that’s needed significant automation for a long time, and we, as in AI, can be a multiplier for these departments, these organizations. It’s a very exciting time — this is one of the hardest problems in B2B.

Jay: I think we’re just going to call it MI — Matt Intelligence — as opposed to AI. But I agree with you, it’s the opportunity for people to start automating things that have been a challenge for a while for many people.

Floyd: This has been super interesting, Jay, and I feel like there’s a whole other data layer to dig into — I’d love to have you back to do that.

Jay: To part two, Floyd.

Floyd: Yeah, part two — and then part three, we debut your rap alter ego, J-Rocks.

Jay: I’m already working on that.

Floyd: I can handle the branding, don’t worry about it, I’ve got you covered. But before we wrap up part one, can you tell listeners where to find you, and how you might be able to help them on their product information journey?

Jay: I’d love to have people come to inriver.com — there’s a host of resources there, both around best practices and research we’ve done that can be useful as people come up to speed on PIM or want to understand some of the best practices. Also, reach out about the AEO assessments I mentioned earlier — it’s an opportunity to get a quick snapshot of where you stand in the market today.

Matt: Thank you, Jay, it’s been a pleasure. Thanks for tuning in to this episode of Data Versus Commerce. New episodes drop weekly.

Floyd: So if you’re responsible for any part of how products get from a database to a doorstep, subscribe now on Apple, Spotify, or wherever you listen.