Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteChoose Import when refreshed data is acceptable and the model fits your capacity; Microsoft recommends it as the default because it offers in-memory performance and broad modeling flexibility. Choose DirectQuery when data must remain at its source or importing it is impractical, and that source can handle interactive report queries. Choose a live connection when an existing Power BI or Analysis Services semantic model already owns the business logic and governance your report should use.
The right choice depends on more than row count. Compare freshness, response time under realistic load, refresh feasibility, modeling needs, security ownership, and connector and gateway requirements. Hybrid tables, Direct Lake, composite models, or aggregations may also address requirements that would otherwise seem to call for pure DirectQuery.
As an Amazon Associate I earn from qualifying purchases.
What the three connection choices mean
Import: a refreshed snapshot in the model
Import loads selected data into the semantic model’s in-memory cache. Report visuals query that cached data, which typically makes interactions responsive. Changes at the source appear after the model refreshes, so the refresh schedule must match the business need and fit available capacity.
Microsoft’s current guidance is to use Import by default, then choose another approach only when a specific constraint—such as freshness, size, governance, security, or architecture—requires it. See Microsoft’s DirectQuery guidance.
#1 Best Overall
DirectQuery: the model asks the source for data
DirectQuery does not import table data into the semantic model. Visual interactions generate queries against the source, so results can reflect newer source data after a requery. Responsiveness depends on the source, network, and any gateway in the path; queries also add load to the source. Microsoft lists very large data volumes, near-real-time requirements, and federated access as possible reasons to consider it, not as automatic reasons to select it.
DirectQuery is not a guarantee of real-time values: caching and visual refresh behavior affect what a report displays. It also has transformation and modeling limitations. Review the DirectQuery overview and Microsoft’s Desktop guidance before designing around it.
Rank #2
Live connection: a report uses an existing semantic model
A live connection attaches a report to an existing Power BI semantic model, Azure Analysis Services model, or SQL Server Analysis Services model. The report consumes that model rather than creating a local semantic model, so the upstream model remains responsible for its definitions and governance. Some modeling options are unavailable in the connected report.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →In the documented live-connection behavior, the user’s identity is passed to the upstream model for security trimming. A live connection is a report-authoring choice, not another name for DirectQuery: the remote model can itself use different table storage modes. See Microsoft’s live connection documentation and its storage-mode overview.
Choose based on the requirement
| Requirement | Start by considering | What to check |
|---|---|---|
| High interactivity and broad transformation flexibility; periodic refresh is acceptable | Import | Model size, refresh duration, capacity, and how stale data may be. |
| Importing the full dataset is impractical, or a source query is needed for fresher data | DirectQuery | Connector support, transformations, source response time, network or gateway overhead, concurrency, and security. |
| An enterprise model already defines governed measures, relationships, and access | Live connection | Required permissions and whether the report can accept the connected-model authoring limitations. |
| Recent facts need fresher values, but older history can be cached | Hybrid table | Microsoft describes a DirectQuery partition for recent data alongside imported historical partitions. Confirm the design suits the model and workload. |
| The source is a qualifying Fabric lakehouse or warehouse and low-latency reads are needed | Direct Lake | Confirm the applicable Fabric setup and behavior; Microsoft’s decision guide identifies it as an option for large Fabric data with low-latency reads. |
| Repeated queries can use summaries, or remote latency and concurrency are concerns | Import or aggregations over DirectQuery | Check that aggregation design and coverage match the report’s actual queries. |
| Several external sources need to be combined without full ingestion | DirectQuery composite model | Source support, composite-model limits, and cross-source query behavior. |
These are starting points rather than guarantees. Microsoft directs users to each connector’s “Capabilities supported” section to verify mode support: DirectQuery in Power BI.
Check performance before choosing DirectQuery
Microsoft’s published Power BI Desktop guidance recommends that source responses for visuals take five seconds or less. It says responses longer than 30 seconds produce a poor report experience, and a query longer than four minutes times out in the Power BI service. These are Microsoft recommendations and documented service behavior, not measured results for a particular system. The guidance is in Use DirectQuery in Power BI Desktop.
Rank #4
Test representative visuals and filters with the source, network, and gateway configuration the published report will actually use. Include realistic concurrent use and security trimming, not just a quick single-user query. A source that responds quickly in isolation may still struggle when many report interactions arrive together.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Work through the decision in this order
- Set the freshness requirement. Decide whether users can work from a refreshed snapshot or need newer source values after report interactions. Do not equate DirectQuery with an unconditional real-time guarantee.
- Check whether Import is operationally feasible. Confirm model size, refresh duration, capacity, and any gateway or licensing needs against the intended refresh cadence.
- Identify who should own the model. If a governed semantic model already contains the required measures, relationships, and access rules, determine whether reports should consume it through a live connection instead of recreating that logic.
- Verify DirectQuery end to end if it remains a candidate. Check connector capabilities, transformations, credentials, gateway needs, source permissions, response times, and expected concurrency.
- Compare alternatives to a pure DirectQuery design. Assess hybrid tables, Direct Lake where applicable, composite models, and aggregations against the actual workload.
- Confirm change and operating constraints before implementation. Microsoft documents that changing a table from DirectQuery to Import generally cannot be reversed to DirectQuery in Power BI Desktop; specific exceptions exist in web modeling and live editing. Check the current storage-mode documentation before committing.
Security and operations are part of the choice
For Import and DirectQuery, establish how the published model connects to its sources, which credentials it uses, whether an on-premises gateway is required, and how source permissions interact with report access. For a live connection, verify access to the upstream model and its identity and security behavior. Connector support and service setup vary, so check the relevant mode and connector documentation rather than assuming a Desktop connection will work unchanged after publishing.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
