Building Scalable Product Subscription Models with Row and Column Filtering

This guide is written for Producers who want to design scalable data offerings that support multiple target consumers without increasing operational complexity. It is also relevant to Operators responsible for the governance and monetization of their marketplace. It is useful for teams that are beginning to populate their Exchange, as well as those who can now re-structure access in a way that improves governance, eases discovery, and supports commercial outcomes.

Most successful data products serve more than one audience. Internal analysts may require full access to explore and model the data, external partners may only need a regional subset, and executives often need a simplified view designed for dashboards and reporting. Traditionally, supporting these varied requirements has meant creating multiple versions of the same product — a single dataset appearing in several forms: a full version for internal teams, a filtered version for partners, and a restricted dataset created to satisfy governance requirements.

While each version serves a legitimate use case, maintaining separate products introduces unnecessary complexity across all three platform personas.

For Producers, it increases the packaging, maintenance, and governance load beyond what is necessary. Each product must be designed and maintained independently, and the risk of misconfiguration and human error grows with every version.
For Consumers, it creates friction at the point of discovery. Browsing through duplicate or near-identical product listings slows time to insight and can make it harder to identify which product is right for their needs.
For Operators, product sprawl clutters the Exchange and obscures Plan upgrade pathways. When it is unclear to Consumers how to move from a trial to full access, conversion rates suffer and the revenue potential of the marketplace is limited.

Row and Column Filtering addresses this by allowing Producers to define access once at the Plan level and governance is enforced automatically across all consumption points. The result is a model where a single, core product can serve multiple audiences without multiplying packaging, or manual overhead — allowing products to scale without scaling complexity.

Supporting Commercial Outcomes Through Tiered Access

For Consumers, subscription-level filtering simplifies both discovery and usage. Instead of browsing through several similar datasets, Consumers encounter a single product page with clearly defined access options. Each subscription plan communicates the scope of data it provides, helping Consumers self-select the version that best matches their needs — making it more likely they find, understand, and act on what's available. This is critical, given that successful marketplaces rely on Producers who take responsibility for commercial outcomes, not just the technical steps leading to publish. A single product with well-structured subscription plans gives Consumers a clear, navigable entry point to buy data.

Filtered plans also create a natural foundation for trials that convert to paid subscriptions, and by extension, an enhanced commercial model. A trial plan can be configured with time-bound access and row or column restrictions that give Consumers a meaningful but intentionally scoped view of the data — a single region, a limited date range, or a subset of columns — providing enough to validate the product's value before committing to full, paid access.

Critically, the path from trial to full access should be designed deliberately and thoughtfully. Producers should ensure that the product page makes the upgrade journey clear: what additional data, whether rows or columns, a paid Plan unlocks, and how to access it. When this pathway is visible and intuitive, filtered Plans become the mechanism by which products prove value, protect ownership, and scale revenue. The marketplace and its Operators benefit from this model, and with it, the cleaner Exchange that it brings along with it. Less duplication and clearer product hierarchies reduces the friction that Consumers undergo before reaching the point of purchase.

Designing and Framing Plans Effectively

The filters applied create a set of rules that determine whether the consumer receives full access or a filtered subset of the data. When filtering is applied, it is clearly communicated on the product page so Consumers understand that the plan provides a limited view of the dataset. Columns that are excluded appear greyed out in the data dictionary, and row filtering logic can be viewed directly. Producers can also write a filtering description to explain what segment of the data subscribed Consumers will gain access to. This transparency ensures Consumers always understand what data they are working with and what limitations apply.

How plans are named and described has a direct impact on whether Consumers can self-serve confidently. While a plan labelled "Standard Access" tells a consumer very little, one labelled "UK Consumer Data — Operational Metrics (Excludes PII)" immediately communicates scope, geography, and sensitivity. To accelerate time-to-access, Producers should treat Plan names and filtering descriptions as consumer-facing content, as opposed to labels for internal use.

When writing Plan descriptions, Producers should aim to answer three questions from the consumer's perspective: What data will I receive? What is excluded? And is this the right Plan for my use case compared to the others on offer? Filtering descriptions are the primary place to communicate this — use them to describe the specific region, date range, or fields that the Plan provides access to, rather than repeating general information about the dataset.

Aligning Access Offering With Consumer Needs

Row and Column Filtering is most effective when a single dataset must support multiple audiences or use cases. Rather than creating separate products for each group, producers can define subscription plans that align with how different Consumers interact with the data.

When designing these plans, producers should consider three factors: the intended audience, the sensitivity of the data, and the consumer's use case. Audience determines who the data is for, sensitivity determines which fields or records require protection, and use case determines the level of detail required for the data to be useful. Considering these dimensions ensures that plans reflect how the product will actually be used, rather than how the data happens to be structured.

Consider a product built around the previously mentioned UK and Ireland Industry Consumer Data containing fields such as transaction ID, date, amount, and customer attributes. Rather than publishing several versions of this dataset, the producer can create multiple subscription plans:

An Internal Analytics Plan might provide full access to the dataset, allowing data teams to perform exploratory analysis and build models.
A Commercial Partner Plan might restrict rows to specific regions and remove PII fields, allowing external partners to access insights without exposing sensitive information.
An Executive Reporting Plan might expose only high-level metrics or aggregated views designed for dashboards and performance tracking.
A Trial Plan might provide access to a single customer type and a limited date range — enough for a consumer to evaluate relevance before committing to a paid Plan.

Here, each Plan serves a different audience, yet all originate from the same core product. Row and Column Filtering allows Producers to segment access once and leave the platform to enforce those rules consistently across the entire product lifecycle. As new audiences emerge or use cases change, Producers can add, edit, or expire subscription tiers independently — extending the product's value offering without creating new packaging or product tiles on the Exchange.

Consumers receive the version of the dataset they are entitled to — whether accessing it in Delta Sharing, or via Export — without needing to manually filter or transform the data downstream. This reduces manual preparation work, lowers the risk of human error, and empowers Consumers to move from discovery to insight more quickly.

Governance as Producer Protection

Row and Column Filtering is not only a mechanism for meeting data governance requirements, but also for Producer confidence: it is what makes sharing data externally viable in the first place.

When sensitive fields can be excluded at the Plan level and that rule is enforced automatically across consumption points — Delta Sharing and Export — Producers can offer broader access without having to accept greater risk. A Producer who might otherwise be cautious about sharing a dataset externally can do so knowing that PII or sensitive records are unavailable to Consumers on certain plans, regardless of their role or subscription method.

This matters particularly for Organizations operating across regulatory boundaries, or those sharing data with external partners where the scope of access must be defined contractually and enforced technically. Row and Column Filtering provides that technical enforcement layer — not as a replacement for governance policy, but as its operational presence within the platform.

Producers should review filtering rules against their Organization's governance and compliance requirements before releasing a product. Column exclusions should be verified against data classification policies, and row-level filters should be confirmed to reflect the intended scope for each consumer group. Once published, these rules are enforced automatically — but the responsibility for defining them correctly rests at the point of ‘Create a Data Product’.

Conclusion

As data marketplaces grow, the challenge is not simply publishing more data products — it is structuring them in a way that remains clear, governed, and easy to use. Row and Column Filtering allows producers to support multiple audiences from a single core product, replacing duplicated datasets with structured subscription plans.

Access is defined once and enforced consistently across Exports and Delta Sharing. Consumers see exactly the version of the data they are entitled to, while Producers avoid the operational overhead of maintaining duplicative products. Operators benefit from a cleaner Exchange, clearer upgrade pathways, and a commercial model that scales alongside the Marketplace. The result is stronger governance, less operational complexity, and products that grow commercially without growing the burden on the teams that build and maintain them.

Success Checklist

Plans reflect real consumer audiences and use cases, not dataset structure
Filters are applied deliberately, with sensitivity and governance requirements considered
Filters are reviewed against compliance policies before and periodically after release.
Plan names are written for the consumer, communicating scope and restrictions clearly
Filtering descriptions state exactly what data each Plan includes and what is excluded
Trial Plans are scoped to demonstrate value, but not enough to undercut that of full access. Where a trial and paid Plan exist, upgrade pathways are explicit
Plans are updated as use cases evolve, not replaced with new product versions.