There is a pattern I keep seeing on LinkedIn and in conversations with people at other organisations. A team starts a new Power Platform project, someone labels it ‘enterprise’, and suddenly the backend discussion jumps straight to Dataverse. No one questions it. Licensing gets justified in a spreadsheet, premium seats get budgeted, and six weeks later the solution is a list of 4000 records with a status column and three lookup fields. That is the moment I want to push back. For a big chunk of internal work, SharePoint lists for power platform apps are still the right call, and treating that choice as amateur hour is costing teams real time and money.
Why SharePoint Lists Keep Getting Dismissed Unfairly
The dismissal usually comes from two places. First, the word SharePoint carries baggage from the 2013 era that has nothing to do with what Lists actually is in 2026. Second, Dataverse gets sold as ‘the proper enterprise choice’ in every Microsoft deck, so picking anything else feels like admitting you cut corners.
Neither of those is an engineering argument. They are vibes.
In practice, a modern SharePoint list gives you a REST and Graph API, column-level validation, versioning, item-level permissions if you really need them, native integration with every Power Platform connector, and a UI that business users can actually open and fix when something looks off. You get Entra ID authentication for free. DLP policies cover it. The audit log flows into the same Purview tenant everything else does. For an internal request tracker, an onboarding checklist, a small approval queue, or a reference table that three flows read from, that is the whole job.
The Workloads Where SharePoint Is Genuinely the Right Answer
I am not arguing SharePoint wins everywhere. I am arguing it wins for a specific and very common shape of work.
If the app is an internal tool, used by under a few hundred people, with a data volume that will never credibly pass 20 or 30 thousand items in any single list, and a schema that fits cleanly into flat records with a handful of choice and lookup fields, you are in SharePoint territory. Registrations, comms sign-offs, small approval workflows, asset inventories, internal request forms, training trackers, audit checklists. This is the bread and butter of Power Platform inside any large organisation.
Building on a list for that kind of workload is faster in almost every dimension. You spin up the schema in minutes. Makers on your team can read and edit the data directly in the browser without a model-driven app wrapper. Your Power Automate flows use the standard SharePoint connector, which means no premium licence for the thousand people who happen to trigger a flow indirectly. That last point alone saves real budget. I have written before about how licensing shapes architecture choices in enterprise environments, and this is one of the clearest examples.
The Real Limits You Need to Design Around
SharePoint lists are not magic. They have hard edges, and you need to know them before you commit.
The 5000 item view threshold is the one everyone quotes, and it is real, but it is a view and query limit, not a storage limit. A list can hold millions of items. You work around the threshold with indexed columns, filtered views, and delegable queries in Power Apps. If you are careful with your delegation, a list with 50,000 items performs fine for the usage patterns most internal apps actually have.
Throttling is the other one. The SharePoint connector has its own limits, and if you are running a flow that hammers a list in an Apply to Each loop at full concurrency, you will get throttled. Same story I covered in the throttling post. Drop concurrency, batch your reads, and stop treating the list like a transactional database.
Where SharePoint genuinely falls over is relational complexity. More than two or three lookups deep and you are building joins in Power Automate, which is painful to maintain. Row-level security beyond basic item permissions gets ugly fast. Calculated columns are limited. And if two users edit the same item in the same second, you can get conflicts the platform does not resolve gracefully. This is also where why Power Automate is still worth learning in 2026 becomes relevant — knowing how to structure flows around list limitations is a skill that pays off repeatedly.
When to Actually Move to Dataverse
Dataverse earns its licence cost when the workload has specific characteristics. Genuine relational depth with many-to-many relationships you actually traverse. Role-based security that needs to be enforced at the data layer, not the app layer. Business rules and calculated fields that are part of the data contract. Plugins, server-side logic, or integration with Dynamics. Auditing at the field level for compliance. Transactional guarantees where a failed write must roll back cleanly.
If your project has three or more of those, Dataverse is the right choice and the licensing conversation is easy to win because the alternative is building those features yourself and maintaining them forever.
If your project has none of them, and you are about to spend eight weeks justifying premium seats for 400 people so you can store a status column on 4000 records, stop. Build it on a list. Ship it next week. Spend the saved budget on something that actually needs it. And before you wire any intelligence layer into whatever you build, it is worth reading Adding Copilot to Your Power App Is Not the Same as Making It Smarter — the backend choice and the AI layer choice follow the same logic.
The backend choice is an engineering decision, not a status symbol. I write more about these kinds of trade-offs on my LinkedIn if this is the kind of call you are wrestling with right now.
Frequently Asked Questions
When should I use SharePoint lists for Power Platform apps instead of Dataverse?
SharePoint lists are a strong fit when your app is internal, used by fewer than a few hundred people, and will never exceed tens of thousands of items in a single list. Common examples include approval queues, request forms, training trackers, and asset inventories where a flat schema with a few lookup fields covers everything you need.
Why does using SharePoint as a Power Platform backend save money on licensing?
Power Automate flows that connect to SharePoint use the standard connector, which does not require a premium licence. In large organisations where thousands of users might trigger a flow indirectly, avoiding premium connector requirements can add up to significant budget savings compared to a Dataverse-backed solution.
What are the security and governance features available with SharePoint lists?
Modern SharePoint lists authenticate through Entra ID and are covered by your organisation’s DLP policies. Audit logs feed into the same Purview tenant as your other Microsoft 365 services, so governance and compliance requirements are met without additional configuration.
How do I know if Dataverse is actually necessary for my Power Platform project?
If a solution is genuinely enterprise in scale, involving complex relational data, high record volumes, or advanced role-based security requirements, Dataverse earns its place. However, if the data model is a few flat lists with simple lookups and the user base is a small internal team, reaching for Dataverse by default adds cost and complexity without a clear engineering benefit.
Leave a Reply