Skip to content
Tutorials Center
My learning

Topics

Lesson 39 of 56 ยท Visuals and Reports

Drill-through, Tooltips, Bookmarks and Buttons

Make Power BI reports explorable: drill down and drill through, report-page tooltips, bookmarks, navigation buttons and what-if parameters, with Python stand-ins.

Filters and slicers narrow a page. Real exploration goes further: a reader sees a bad number, right-clicks it and lands on a page with the detail behind it; they hover over a bar and get a mini chart; they click a button and the report jumps to a saved view. These features turn a stack of pages into something closer to an app, and none of them needs any code. They are settings on pages, visuals and buttons.

This lesson covers five tools: drill down, drill-through, report tooltips, bookmarks with buttons, and what-if parameters. Power BI Desktop does the actual work, so the steps are described in text, with one DAX snippet shown read-only. The Python stand-ins mirror what each feature does to the data: grouping at each hierarchy level, pulling a detail table, taking snapshots of a report's state and testing scenarios. If slicers are new to you, read Filters, Slicers and Sync first.

Five ways to go deeper

Drill down Next level of a hierarchy.
Drill through Jump to a detail page.
Tooltip page A mini report on hover.
Bookmarks Saved views of the report.

A button can trigger most of these.

1 / 2

Drill down and drill up

When a visual's axis holds a hierarchy (Category > Product, or Year > Quarter > Month), the visual header shows drill icons. You create hierarchies in the Model view (see Model View, Hierarchies, Formatting and Display Folders) or by dragging several fields into the axis well.

  • Drill down (the down-arrow icon) turns drill mode on. Click a bar, and the chart replaces its axis with the next level, filtered to that bar.
  • Go to the next level (double down arrows) shows the next level for all items at once.
  • Expand to the next level (forked arrow) keeps the current level and adds the next below it, so labels become "Bikes - Road".
  • Drill up (up arrow) goes back.

Drill down changes only that visual. It does not filter the rest of the page, which sets it apart from cross-filtering. Use it when a chart has an obvious natural breakdown and you do not want to build a separate page for it.

Drill through

Drill through takes the reader from a summary to a page about one item. You build the target page once and every summary visual can use it:

  1. Add a page, for example "Product detail", and design it with the detail visuals (orders, customers, trend).
  2. In the Visualizations pane, drag the field that identifies the item (such as Product[ProductName]) into the Drill through well.
  3. Optionally switch Keep all filters on, so the page also respects the filters from the source page.
  4. Power BI adds a Back button automatically at the top left.

Now readers right-click a bar, a table row or any data point that uses that field, choose Drill through and pick the page. The detail page opens filtered to that item. You can add more than one field to the well (the page then appears for either), and you can drill through across reports when both are in the same workspace by enabling Cross-report.

flowchart LR
    A["Summary page<br/>sales by product"] -- "right-click a bar,<br/>Drill through" --> B["Detail page<br/>filtered to that product"]
    B -- "Back button" --> A

The detail page is built once and reads the product the reader chose as a filter.

The stand-in below does the same thing to a table. The first part is drill down: the same measure at two levels of a hierarchy. Then comes drill through, which filters the detail rows to one product, and a tooltip-style summary of one category.

Python Drill down, drill through and a tooltip

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Drill down never changes the data, only the level of the grouping: bikes total 3,270 and split into road (1,850) and trail (1,420). Drill through adds a filter on the product, so the detail page lists the two road bike orders for 1,850. Both use the same filter-context idea as slicers.

Report page tooltips

A tooltip appears when a reader hovers over a data point. The default shows the field values; a report page tooltip shows a whole small page, with its own visuals, for the item under the mouse. To create one:

  1. Add a page, open Format > Page information and switch Allow use as tooltip on. In Page settings choose the Tooltip page size (a small canvas).
  2. Drag the field the tooltip should respond to into the Tooltip well (for example Product[ProductName]), or leave the default so it accepts the visual's own fields.
  3. Design the page with two or three small visuals: a trend sparkline and a few cards.
  4. On the visual that should show it, open Format > General > Tooltips, set the type to Report page and pick the page.

The tooltip page is filtered to the hovered item, the same way a drill-through page is. Keep it light, because the page runs as a query on every hover; a tooltip with five heavy visuals makes the report feel slow. For simpler needs, you can just drag extra measures into the visual's Tooltips well and the default tooltip lists them.

Bookmarks

A bookmark captures the current state of a report page so you can return to it later. Open View > Bookmarks and choose Add. A bookmark can remember:

  • Data: filters, slicer selections, sort order and the drill location.
  • Display: which visuals are shown or hidden, spotlight and focus mode.
  • Current page: the active page, so the bookmark can jump there.
  • All visuals or selected visuals only. Selecting only some visuals lets a bookmark change part of a page and leave the rest alone.

A bookmark does not store data. It stores the settings that produce a view, so a bookmark saved with "year 2024" shows the latest data for 2024 after the next refresh.

Bookmarks combine with the Selection pane to make toggles. A common pattern is to put two charts on top of each other (revenue and profit), give each a name in the Selection pane, and make two bookmarks that show one and hide the other. Two buttons then switch between them. Another is a "Reset" bookmark with no filters, which a button can apply.

You can also show bookmarks as a Bookmark navigator visual (Insert > Buttons > Navigator > Bookmark navigator), which builds a row of tabs from your bookmarks automatically.

Buttons and navigation

A button (Insert > Buttons) is a shape that runs an action when clicked. The main actions are:

Action What it does
Back Returns to the previous page, or the page you drilled through from
Bookmark Applies a saved bookmark
Page navigation Opens a given page; also a Page navigator visual builds the tabs for you
Drill through Goes to a drill-through page with the current selection
Apply all slicers / Clear all slicers Controls slicers together
Q&A Opens the question-and-answer experience
Web URL Opens a link

Buttons can have different looks for default, hover, pressed and disabled states, and a conditional tooltip text. A navigation menu built from a page navigator is easier to maintain than hand-drawn buttons, because it updates itself when you add or rename pages.

The stand-in below treats a bookmark as a snapshot of selected parts of the report state, and a button as an action that applies a snapshot or changes the page. It shows that a bookmark captures settings, and that choosing what it captures is what decides what it changes.

Python Bookmarks as saved state

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Applying "North 2024 profit" restored the year, region and chart but left the reader on the Detail page, because that bookmark was captured without the current page. In Power BI the same effect comes from turning Current page off in the bookmark's options. A bookmark that includes everything would also pull the reader back to the page where it was created.

What-if parameters

A what-if parameter lets the reader try a scenario by moving a slider. Choose Modeling > New parameter > Numeric range, give it a name, a minimum, maximum and step, and keep Add slicer to this page ticked. Power BI creates a small table with a column of values and a measure that reads the selected one, as in:

DAX
Discount = GENERATESERIES(0, 0.5, 0.05)

Discount Value = SELECTEDVALUE('Discount'[Discount], 0)

You then use [Discount Value] in your own measures:

DAX
Profit After Discount =
SUMX(Sales, Sales[Quantity] * (Sales[UnitPrice] * (1 - [Discount Value]) - Sales[UnitCost]))

This is the disconnected-table pattern from Common DAX Patterns, built for you. Moving the slider changes the measure, so every visual that uses it updates live. The stand-in tries the same measure for several discount values.

Python A what-if scenario

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Profit falls from 770 at no discount to 224 at a 20% discount, losing 136.50 with every extra 5 points, so a slider shows quickly how much margin a discount costs. The stand-in uses the same formula as the measure, but the slider and the live refresh of the visuals only exist in Power BI.

Using these features well

Interactivity is easy to overdo. Give each feature a clear job: drill down for natural hierarchies, drill-through for "show me the details", tooltips for a quick look, bookmarks for alternative views and buttons for navigation. Tell readers what is clickable (a short hint such as "Right-click a product for details") because hidden features go unused. Test the report from the reader's point of view, including the Back button. Once the report is ready, you can add visual cues for status, which is the subject of Conditional Formatting, KPIs and Cards.

Check your understanding

Quick check ยท 5 questions

of 5 answered You scored / 5 ยท perfect!

  1. 1. What is the main difference between drill down and drill through?

    Drill down changes the grouping level of one visual. Drill through opens a separate page, filtered to the chosen item.
  2. 2. What must you set up to make a page usable as a report tooltip?

    The page needs Allow use as tooltip, and the visual's Format > General > Tooltips settings must point to it as a Report page.
  3. 3. A bookmark is saved with the Current page option off. What happens when a button applies it?

    Turning off Current page means the bookmark does not move the reader. It still applies the data and display settings it captured.
  4. 4. What does a bookmark store?

    Bookmarks save the state that produces a view, not the data, so they show current data after a refresh.
  5. 5. What does a what-if parameter create?

    Power BI builds a series table, a SELECTEDVALUE measure and a slicer, which you then use inside your own measures.

Summary

  • Drill down changes the hierarchy level inside one visual; drill through opens a detail page filtered to the chosen item (a Back button is added for you).
  • A report page tooltip is a small page shown on hover; allow it in the page settings and choose it in the visual.
  • Bookmarks save settings (data, display, current page), not data; combine them with the Selection pane and buttons to toggle views.
  • Buttons run actions: back, bookmark, page navigation, drill through, slicer control, Q&A and links.
  • What-if parameters create a series table, a SELECTEDVALUE measure and a slicer for scenario analysis.

Next: Conditional Formatting, KPIs and Cards.