A slow Power BI report is almost always a data model problem wearing a Power BI costume. The usual culprits, and the order to fix them in.

Your Power BI report is slow. Power BI is probably not the problem.

When a report drags, the instinct is to blame Power BI, the network, or the capacity, and to fix it by buying something bigger. That hides the problem for a while and charges you monthly for the privilege. A slow report is almost never a Power BI problem. It is a data model problem wearing a Power BI costume. The tool is fast. The model behind it usually is not, and the visual is simply where you feel the pain.

This connects with our Power BI service and, separately, with Business Central Runs Your Business. It Should Not Run Your Reports.

The reassuring part is that the causes are well understood and the fixes are unglamorous. You can almost always make a report quick again without buying a bigger anything.

The usual suspects

One cause sits far ahead of the rest: a wide, flat table with everything in it, instead of a clean star schema of facts and dimensions. It feels simpler to build, which is why it is everywhere, and it forces the engine to work far harder and compress far worse than it should. It is the single biggest drag on performance in the wild, and the fix is to model it properly.

After that, a familiar cast. Too many high-cardinality columns, the long text keys, the free-text notes, the timestamps down to the second, because the engine that makes Power BI fast relies on repetition it can compress, and those columns give it none. DirectQuery used by default or out of caution, sending a fresh query to the source for every click, when most reports should import and would be far quicker for it. Calculated columns doing a measure's job, computed and stored for every row whether anyone looks or not. And twenty visuals crowded onto one page, each firing at least one query, the slowest holding up the rest.

The decision with the biggest single effect

Choosing the storage mode does more for speed than almost anything else, so make it deliberately rather than by default. Import for most things. Direct Lake when your data already sits in a Fabric lakehouse and is genuinely large. DirectQuery only when you truly need live data or the volume rules import out. A great many slow reports are simply on the wrong mode because someone picked it once and never revisited it.

Work from cause to symptom, not the other way round

There is an order to this, and following it means you rarely have to guess. Measure first, with Performance Analyzer, so you know which visuals and queries are actually slow rather than assuming. Then look at the model size and find the heavy, high-cardinality columns. Then check the storage mode suits the data. Then move logic out of calculated columns and into measures. Then lighten the page. In the great majority of slow reports the cause is found in the first two steps, and the cure costs nothing but a rebuild.

Sometimes the volume is genuinely real and the model is already clean. Then the tools change: aggregation tables, incremental refresh, composite models. They are worth reaching for once the basics are right, and a waste of effort before. Most slow reports never get that far, because the basics fixed them.

Before you ask for more capacity

So next time a report is slow, do not open the procurement conversation first. Run Performance Analyzer and look at the model. The honest answer is usually sitting in the first two steps, and it costs a rebuild rather than a bigger monthly bill. Diagnosing and fixing this properly is everyday Power BI work for us, often as part of a wider Microsoft Fabric estate, and the full diagnostic, with the tools for each step, is in our Why Your Reports Are Slow guide. The bigger engine always feels like the safe fix. It is usually just an expensive way to avoid the real one.

If any of this sounds familiar, talk to us about your data.

Related reading

• What is worth adopting from a Power BI release, and what to ignore
• Three jobs, one dashboard, none of them done well
• Power BI or Business Central’s built-in reporting?