The Expedition: A Manufacturer’s Guide to Getting Product Data Right

Hiram Bingham didn’t wander into the Peruvian Andes and stumble across Machu Picchu. He planned an expedition. Most manufacturers implementing a PIM skip every step he took, and the result is entirely predictable.

Most manufacturers who implement a PIM do so with a clear conviction that the hard part is selecting the platform. They know their product knowledge exists somewhere inside the organization. They know it needs to be structured and made accessible for their digital channels. And then they buy a platform, hand the project to IT, and expect the data to materialize.

Without a team built for the work, without a methodology for finding and evaluating what gets surfaced, and without a realistic picture of the resources required, the result is predictable: implementations that stall mid-project when the actual scope of the excavation becomes clear, timelines that stretch from months to years, and PIM systems that go live with a fraction of the content that was planned.

This paper is about how to run the expedition properly.

How Manufacturers End Up in a PIM Conversation

Understanding what triggered the PIM decision tells you almost everything about how clearly a manufacturer has diagnosed their own problem.

For most manufacturers, the trigger is external pressure. A large distributor starts putting requirements on the data they receive. They want content mapped to their taxonomy. They want attributes that meet their quality standards, not a generic spreadsheet export that the manufacturer has been sending to every channel partner regardless of size or digital sophistication.

So the manufacturer enters the PIM evaluation thinking: we have a syndication problem. We need to deliver our data in different formats to different channel partners. That is a real problem and worth solving. It is not the primary problem. The moment a manufacturer begins implementing a PIM to solve a channel formatting challenge, they encounter the thing nobody warned them about: before you can syndicate structured product content to anyone, you have to have structured product content to syndicate.

What the Excavation Problem Actually Looks Like

The knowledge a manufacturer needs to populate a PIM completely and accurately is almost always present somewhere inside the organization. That is the good news. The bad news is what “somewhere” actually means in practice.

It means a legacy ERP system containing product records entered by people who are no longer there, in formats designed for order management rather than customer-facing content. It means CAD files with precise dimensional and performance data that has never been translated into customer-facing attribute language because the engineering team never needed to express it that way. It means specification sheets from three acquisitions ago, sitting in a shared drive nobody outside the original business unit knows exists.

And it means tribal knowledge: the applications engineer who has been with the company for twenty-seven years and can tell you exactly which fastener configuration is appropriate for a marine-grade stainless enclosure under thermal cycling stress, and who has never been asked to write that down in a form a website can use because until now, customers got that answer by calling him directly.

Discovery

Finding where product content actually lives across the organization. This requires a systematic audit of every system, every shared drive, every documentation repository, every team that touches product information in any form. Content that nobody thought to look for cannot be captured.

Extraction

Getting the content out of its source format and into a form that can be evaluated for PIM input. An engineering drawing is not a structured attribute. A PDF installation guide is not a digital asset linked to a product record. Getting from source format to PIM-ready content requires significant work at every step.

Validation

Determining whether what has been extracted is accurate, current, and appropriate for external use. Specifications get superseded. Products get modified without the documentation being updated. Every piece of extracted content requires someone who understands the product to confirm it is right before it goes into the PIM.

Why IT Ownership Fails, and What to Do Instead

The organizational reality of most manufacturer PIM implementations is that IT owns the project. This makes a certain kind of sense: a PIM is a technology platform, and technology projects belong to IT. It is also the single most reliable predictor of an implementation that stalls in the content phase.

IT is the right team to manage the platform configuration, the system integrations, the data architecture, and the technical infrastructure. IT is the wrong team to determine which legacy system contains the authoritative version of a product specification, evaluate whether a document extracted from a 2009 engineering archive reflects the current product design, or decide which attributes are essential for a customer trying to specify a product for a particular application.

Those judgments require product knowledge. And product knowledge lives with the people who own the products: the product managers, the merchandisers, the applications engineers, the people responsible for the product lifecycle from development through end-of-life.

IT manages the system. The business determines what goes into it and confirms that what went in is right. This is not a subtle distinction. It is the difference between an implementation that produces a populated PIM and one that produces a technically configured PIM with a content gap the organization will spend the next three years trying to close.

Making this work in practice requires resolving a question most organizations avoid until it becomes unavoidable: who, specifically, is accountable for product content quality? Not which department in general. Which person, with which authority, with which performance expectations attached to the outcome.

In manufacturers that have gotten this right, the answer is usually a role that sits at the intersection of product management and digital commerce: a product content manager, or a digital commerce product owner whose job is explicitly the quality and completeness of product information as a digital asset. This role did not exist at most manufacturers a decade ago. At the manufacturers advancing fastest digitally, it exists now.

The Resource Reality

Before any manufacturer begins a PIM implementation, one question needs an honest answer: where are the people coming from?

The instinct in most organizations is to treat the PIM project as a workstream that existing teams absorb alongside their current responsibilities. Nobody’s headcount changes. The project gets done as a parallel track. This does not work. It has never worked, and the reason is not a project management problem. It is a logic problem.

Consider the analogy of opening a new distribution center. If a company opens a facility in Toledo, it cannot staff it by pulling people from its operation in Columbus. Columbus still has orders to fulfill. Columbus still has customers to serve. The work in Toledo is entirely new work serving an entirely new purpose, and it requires entirely new resources, not a reallocation of existing ones.

The Honest Business Case

The cost of the content work needs to be in the business case before the platform decision is made, not discovered six months into implementation when the gap between what was planned and what is actually required becomes impossible to ignore. A digital operation is a new branch. Staff it as one.

The Emerging Technology Layer: AI-Powered Content Discovery

Until recently, every step of the excavation process was manual. Someone had to know where to look, go find the content, extract it from its source format, evaluate whether it was accurate, and structure it for PIM input. At scale, for a manufacturer with a large, complex catalog and decades of accumulated product documentation, that manual process was the primary constraint on how fast a PIM could be populated.

That constraint is beginning to shift. A new category of AI-powered tools is emerging that approaches the internal content discovery problem the way a search crawler approaches the web: by systematically traversing internal networks, identifying content sources, and surfacing structured data that can populate PIM attributes. Platforms like Anglera, MerchKit, DataX, and Rastro represent early entrants in this space. The category is evolving, which means capabilities and integration maturity vary considerably.

The critical point for any manufacturer evaluating a PIM today: this capability is not yet native to most PIM platforms. It requires integration with a separate tool. A manufacturer who selects a PIM without asking how it integrates with content discovery tools is making a platform decision without understanding how the hardest part of their implementation is going to get done.

What Becomes Possible When the Source of Truth Exists

Once a manufacturer has built a genuine, structured, validated source of truth for their product content, two categories of value become accessible that were entirely out of reach before.

The first is channel syndication at scale: the ability to deliver product content to any channel partner in whatever format that partner requires, without starting from scratch for each one. The content exists in a clean, structured form. The syndication layer translates it into whatever the destination requires.

The second is channel enablement at scale. A manufacturer with a comprehensive PIM can make the same depth of product knowledge available to every channel partner, regardless of the partner’s size. The small regional distributor whose sales team has no deep expertise in the manufacturer’s products can access the same technical content, the same application guides, the same comparison tools that the largest national distributor gets. A distributor whose team has access to rich application content will sell on value. The manufacturer who enables that sells more product at better margins through the same channel relationships.

Questions to Ask in a PIM Evaluation

What happens before data is in your system?

Every vendor can show the syndication output. The real test is upstream: document parsing capabilities, connectors to legacy ERP, PLM, and CAD systems, and how other manufacturers have extracted content from unstructured internal sources. If the answer pivots immediately to the import template and syndication dashboard, the platform was built for translation. The manufacturer excavation problem is not where its design energy went.

How do you integrate with content discovery platforms?

Ask by name: have you integrated with Anglera, MerchKit, DataX, Rastro, or similar platforms? A vendor who hasn’t thought about this question doesn’t have a credible answer to how a manufacturer with a large, complex catalog is supposed to populate their platform efficiently.

Who owns the data in your most successful manufacturer implementations, IT or the business?

Push past the generic answer. Ask for specifics. The implementations that work have business owners, not just IT project managers accountable for go-live dates.

Who is a reference customer who solved the internal data consolidation problem, not just the channel syndication problem?

If the vendor struggles to produce that reference, the manufacturer excavation use case is not where they have built their strongest capability.

The Expedition Debrief

Hiram Bingham’s 1911 expedition didn’t produce a finished result. It produced a map: the first systematic documentation of what was actually there, which made every subsequent phase of understanding Machu Picchu possible. That is the right mental model for a manufacturer PIM program. The first expedition surfaces the knowledge that has been inside the organization all along, but was never organized for digital use.

The syndication, the channel enablement, the digital experience differentiation, all of it builds on what the expedition surfaces. The manufacturers furthest along digitally are not the ones who found a shortcut through the excavation phase. There isn’t one. They’re the ones who planned the expedition properly, with the right team, the right methodology, the right resources, and a clear understanding that getting the content into the system is the work that makes every other investment pay off.