Your PIM can be technically live and still operationally dead. It may be the case that the integrations run, the User Interface loads, the licences get renewed, but the fact is, the business is still shipping product content through manual correction and completion of spreadsheets (usually through endless email back-and-forth).
When this is the state of affairs, ownership of product data usually defaults to the IT team because, after all, PIM looks like software, doesn’t it? So, tickets go to the service desk, success gets measured in uptime alone, but in the meantime, the real failure mode remains untouched and largely unnoticed: There’s no-one accountable for whether the product data being used is actually sellable.
Gartner, the PIM industry analyst, and ‘scorer’ has its own market definition which makes the point plainly: PIM exists to enable product, commerce, and marketing teams to create and maintain an approved, shareable version of rich product content for multichannel use. That is a business operating model, not an IT infrastructure deliverable
The problem: Who owns the data?
A PIM system sits in your tech stack with its APIs, connectors, permission models, environments, and version updates and patches. As a consequence, businesses generally assume it ‘belongs’ to IT. Moreover, IT accepts it because they’re clearly the only team which has a clear mandate to run systems.
But the genuine value of a PIM solution doesn’t come from running ‘the system’ alone. Rather, it comes from resolving product meaning into structured data which your downstream sales channels can trust. If ownership lies in the hands of IT, there are four predictable behaviours which emerge:
- Uptime replaces usability. The question becomes “Is the job running?” rather than “Is the data ready and usable for selling products?”
- Change simply turns into an IT backlog. New attributes, channel rules, taxonomy changes, or adjustments to workflow are merely more tickets competing with security and infrastructure work.
- The business side disengages. eCommerce, merchandising, and marketing teams fail to take responsibility for any outcomes because they’re treated as ‘requesters’ rather than operators – they feel like they have no agency.
- Shadow processes persist. Teams resort to using ‘parallel’ spreadsheets because the PIM itself isn’t actually the fastest way to syndicate product information.
In the end, none of this is a failure of the technology. It’s simply a lack of data ownership.
The core insight: PIM success is powered by data teams
PIM operations involve decisions, not configuration. Namely, decisions about:
- what an attribute means (and what it does not mean)
- what ‘complete’ looks like per category and per channel
- which values are valid (such as lists, units, formats)
- who can approve what, and on what basis with what criteria
- what changes are permitted without causing downstream feeds to break
Those decisions should belong to the people who are closest to product and channel performance:
- Product data managers
- Category/Merchandising
- Content operations
- Data governance
These people are the ones who understand the commercial consequences of bad data:
- Delayed launches
- Suppressed search visibility
- Rejected marketplace feeds
- Increased returns
- The burden on customer service
All the above aligns with the reality which Gartner highlights regarding intended PIM users: A PIM is designed primarily to support merchandising, marketing, creative and operational roles across the organisation (and often external business partners), not just your technical administrators.
The clean split that organisations avoid making
Many merchants tend to blur two kinds of ownership because as they see it, both “touch the PIM.”
In fact, ownership responsibilities need to be separated:
IT should own:
- platform availability, environments, and security
- integration health and monitoring
- identity and access administration
- vendor management and release governance
Product data teams should own:
- attribute definitions and standards
- taxonomy and classification decisions
- data quality thresholds (per channel because they have unique requirements)
- workflow rules and approvals
- Standards regarding supplier data onboarding (what arrives, in what format, at what level of quality, where it goes when it is ingested, and what is done to it if it doesn’t reach standards)
If we mix those responsibilities, the PIM becomes a kind of ‘political buffer.’ Everyone washes their hands and blames someone else. Data quality becomes “an issue with the system.” User adoption becomes a top-down “training area.” But in the real world, both these examples are issues which impact on ownership.
The ownership test
If you want to know whether your PIM is operating as a product data capability or an IT system, there’s one question to ask yourselves:
Who arrives at work responsible for whether the product data we use in our channels is actually selling our products?
If the honest answer is “IT” (or “Er…No-one”), then you’ve found the limitation holding back commercial performance. IT keeps the oil flowing along the pipelines, but that oil is unrefined – The pipeline isn’t equipped to author product truth, especially not at scale. IT team members aren’t incentivised to argue about attribute semantics, channel nuance, or how much enrichment effort the business can afford per SKU.
What do ‘data-driven operations’ change in practice?
Shifting operational ownership to data teams is far from just re-organising for the sake of it. It fundamentally changes the economics of maintaining product data quality levels.
- Governance becomes an operational discipline. Quality is no longer an abstract aspiration. It becomes a routinely-applied set of thresholds, checks, and approvals owned by named people. (and this is exactly the post go-live reality which the business needs to address in order to prevent value from inexorably dropping over time.)
- Work moves from incident tickets to workflows. Data operators can evolve the schema and rules at the pace the market changes rather than having to wait for urgent sprint cycles.
- Data quality becomes a commercial metric. Completeness and consistency aren’t just treated as ‘data hygiene’ factors, but as key inputs for time-to-market and conversion optimisation.
- Adoption follows naturally from the right fit. When the PIM reflects how teams actually ship products, the tendency of user behaviour to lean towards bypassing the PIM falls away because now, the work is easier done inside the PIM platform than outside it.
The mismatch keeping PIMs in low-performance purgatory
For our experience with many merchants of all types and sizes, weak PIM adoption isn’t a people problem but a fundamental structural mismatch:
Ownership sits with the team which knows how to run software, while value depends on the teams which can sustain product meaning as governed data.
Unless the business remedies that mismatch, it’ll end up running a technically stable PIM alongside (and often separate from) a commercially unreliable product data operation. It’s long-term operational drag disguised as ‘change implementation.’
PIM Discovery call
If your PIM is live but it gets treated as little more than a sterile IT ‘asset’, reach out to us today at Start with Data and book your PIM Discovery call. We’ll map the current ownership split, identify where decisions are being avoided or ticketed, and pinpoint the specific operating-model constraint that’s restricting your user adoption and genuine engagement.