Removing custom hierarchy layers from bar graph with custom visuals.

[mds1234]

I'm creating a bar graph that shows and uses a custom fiscal date hierachy. The date hierachy is Year, Period(1-12), Period Week (1-4), Period Day (1-28). I created a bar chart and using visual calculations to create rolling totals by week. Am I able to remove the third X-axis (week) from this graph? The business user would understand that the bars reset after each week. It just looks very clunky right now. I only want to show the day, period, and year axis labels, while using the week for the running sum.

[Ritaf1983 · Accepted solution]

Hi @mds1234 

From a logical standpoint (not just a technical one), it's not really possible to present a weekly accumulation if the user can’t see any reference to weeks. The moment you remove the week layer from the visual, you're essentially removing the meaning behind the aggregation. Yes — the DAX can still calculate based on fiscal weeks in the background, but for the end user, the result just becomes a confusing jump in values every 7 days without any visible reason. The context is lost, and so is the point of the aggregation.

What you can do instead is rethink the hierarchy structure:

• If you want to reduce clutter, you can build a simplified hierarchy like:

  • Year (top level)

  • Week in Year (as a single, flat level underneath)

• Or if monthly breakdown is important:

  • Year

  • Then a concatenated label like "W-X / M- Y"

• Another approach is to create a custom level like "Fiscal Week Label" that conveys week info compactly (e.g., “2024-Wk03”) without requiring a full hierarchy of Year > Period > Week.

This way, you're still giving the user the semantic clue of what’s being aggregated, even if the layout is simplified.

---

Also, regarding your choice of visual: if you’re showing a column chart, this becomes even more problematic.

Column charts are processed as isolated values — the brain doesn’t interpret them as part of a continuous sequence. If you're showing a cumulative or rolling value, it won’t be perceived correctly. The user will likely interpret each bar as a standalone number and try to compare individual bars rather than recognize a progression.

This gets even worse with clustered bar charts, where the only comparison the user can naturally make is within each group. The horizontal spacing breaks the sense of continuity, and any trend over time is harder to perceive due to the visual gaps and grouped layout.

If you want to show rolling or cumulative trends, a line chart will always be more effective — it's more aligned with how we perceive continuous data over time.

Let me know if you want help restructuring the hierarchy or building a calculated column to generate a cleaner Week label.

If this post helps, then please consider Accepting it as the solution to help the other members find it more quickly.

[v-ssriganesh]

Hi @mds1234,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[v-ssriganesh]

Hi @mds1234,
I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions. If my response has addressed your query, please accept it as a solution and give a 'Kudos' so other members can easily find it.
Thank you.

[rohit1991 · Accepted solution]

Hi @mds1234 ,

Yes, you can absolutely simplify the appearance of your bar graph by removing the week level from the X-axis labels, while still using it in the background for your rolling weekly calculations. In Power BI, the key is to remove the "Week" level from the axis hierarchy shown in the visual, but still keep it available in the DAX measure driving your rolling total.

Here’s how to approach it:

• Keep the full hierarchy (Year > Period > Week > Day) available in your date table and DAX logic.

• In the visual’s X-axis, remove the Week field—only include Year, Period, and Day.

• In your rolling total measure, continue to calculate using Week logic (e.g., by using Period Week in your DAX with FILTER, EARLIER, or RANKX).

• Power BI will still calculate the weekly rolling totals correctly because the DAX is independent of what’s shown on the X-axis.

This way, your chart looks cleaner—only showing Year, Period, and Day—but still functions exactly as needed, with weekly resets in the totals. 

Passionate about leveraging data analytics to drive strategic decision-making and foster business growth.

Connect with me on LinkedIn: Rohit Kumar

[v-ssriganesh]

Hi @mds1234,

Thank you for posting your query in the Microsoft Fabric Community Forum, and thanks to @Ritaf1983 for sharing valuable insights.

Could you please confirm if your query has been resolved by the provided solution? If so, please mark it as the solution. This will help other community members solve similar problems faster.

Thank you.

[Ritaf1983 · Accepted solution]

Hi @mds1234 

From a logical standpoint (not just a technical one), it's not really possible to present a weekly accumulation if the user can’t see any reference to weeks. The moment you remove the week layer from the visual, you're essentially removing the meaning behind the aggregation. Yes — the DAX can still calculate based on fiscal weeks in the background, but for the end user, the result just becomes a confusing jump in values every 7 days without any visible reason. The context is lost, and so is the point of the aggregation.

What you can do instead is rethink the hierarchy structure:

• If you want to reduce clutter, you can build a simplified hierarchy like:

  • Year (top level)

  • Week in Year (as a single, flat level underneath)

• Or if monthly breakdown is important:

  • Year

  • Then a concatenated label like "W-X / M- Y"

• Another approach is to create a custom level like "Fiscal Week Label" that conveys week info compactly (e.g., “2024-Wk03”) without requiring a full hierarchy of Year > Period > Week.

This way, you're still giving the user the semantic clue of what’s being aggregated, even if the layout is simplified.

---

Also, regarding your choice of visual: if you’re showing a column chart, this becomes even more problematic.

Column charts are processed as isolated values — the brain doesn’t interpret them as part of a continuous sequence. If you're showing a cumulative or rolling value, it won’t be perceived correctly. The user will likely interpret each bar as a standalone number and try to compare individual bars rather than recognize a progression.

This gets even worse with clustered bar charts, where the only comparison the user can naturally make is within each group. The horizontal spacing breaks the sense of continuity, and any trend over time is harder to perceive due to the visual gaps and grouped layout.

If you want to show rolling or cumulative trends, a line chart will always be more effective — it's more aligned with how we perceive continuous data over time.

Let me know if you want help restructuring the hierarchy or building a calculated column to generate a cleaner Week label.

If this post helps, then please consider Accepting it as the solution to help the other members find it more quickly.