How do I embed an interactive chart in PowerPoint or Google Slides?

Neither application has a field where an iframe goes. Google Slides can insert a chart linked to a Google Sheet, which refreshes when somebody clicks an Update button rather than on its own. PowerPoint's one documented route to live data on a slide is Microsoft's Power BI add-in, which needs a Power BI license and a tenant that allows the add-in.

For anything outside that, the working pattern is a static image frozen at a stated date with a link to the interactive chart. In a board pack, that is not a compromise. It is the right answer, for reasons that have nothing to do with what the software supports.

What Google Slides actually supports

Insert, then Chart, then From Sheets, then pick a spreadsheet that already contains a chart. Leave the link to the spreadsheet in place and the object keeps a connection to the sheet.

Google's own documentation is specific about what that connection does. When the source data changes, an Update button appears in the top right corner of the chart, and somebody has to click it. There is a Tools, then Linked objects, then Update all path for doing it in bulk. Nothing refreshes by itself while the deck sits in a drive, and nothing refreshes while you are presenting.

There is no web embed feature in Slides at all. The linked objects documentation covers charts, tables and slides from other Google files, and says nothing about external pages, because there is nothing to say. A URL on a slide is a hyperlink, not a frame.

The linked-sheet route is genuinely useful in one case: the numbers live in a sheet somebody maintains, and the deck is rebuilt each quarter. It is a poor fit the moment the chart needs anything the Sheets chart editor cannot draw, which for a yield curve or an underwater drawdown curve is most of the time.

What PowerPoint actually supports

Microsoft documents exactly one way to put live, interactive data on a PowerPoint slide, and it is the Power BI add-in. Within its own boundaries it does what it says: you add a report page or a single visual, the data stays current, and you can filter and drill into it while presenting without leaving the deck.

The requirements are the part the tutorials skip. Office version 2312 build 17126 or higher. Access to the Office add-in store, or the add-in deployed centrally by an administrator. An active Power BI account with at least a Fabric Free license, plus a Power BI Pro license if the report is not on Premium or on F64 or larger capacity. The optional connected experiences setting has to be enabled by an Office administrator. WebView2 has to be present, which it is on recent Windows.

Then the limits. Report pages and individual visuals only, so dashboards and paginated reports are out. Business-to-business scenarios are not supported, which matters the moment the audience sits outside your organization. Microsoft notes that the ribbon button appears in the Windows desktop application, not in PowerPoint for the web and not on a Mac. In sovereign clouds it is available only as an administrator-deployed add-in.

One line in that documentation is worth reading twice, because it decides whether the add-in solves your problem or creates one. The data stays in Power BI and keeps Power BI's permissions, row-level security included. That is correct behavior and good security. It also means the slide is empty for anybody who cannot sign in and reach the report, so a deck emailed to an investor is a deck with a hole in it.

The Web Viewer add-in, and why the tutorials recommending it are out of date

Search this question and most of the results tell you to install an add-in called Web Viewer, paste an HTTPS URL, and get any web page on a slide. Treat that advice carefully. On Microsoft's own Q and A forum, the volunteer moderators answering exactly this question say the add-in has been killed off with no replacement, and the workaround they recommend is a screenshot of the page plus a hyperlink for anyone who wants the live version. That is community moderation rather than a published deprecation notice, so the honest summary is that it is not something to build a recurring board pack around.

There is a second reason it was never a good fit for charts. An add-in renders inside a frame, so a chart host that refuses to be framed shows nothing, exactly as it would inside Google Sites or SharePoint. And a PDF is a static file, so whatever the frame was showing becomes, at best, a picture of the moment the export ran.

Why a live chart in a board pack is a liability

Now the part that matters more than the mechanics, and that none of the tutorials raise, because they are not written for people whose decks are documents of record.

A board pack, an investment committee paper and a quarterly investor update are all minuted. Somebody will return to page eleven in six months and expect the number on it to be the number that was discussed. A chart that re-renders when the file is opened does not do that. It shows whatever the data says on the day the reader opened it, silently, with no indication that it ever said anything else. The meeting agreed one thing, the document now says another, and nobody can tell which version was in the room.

The objection does not apply to a chart you are presenting from live, where currency is the whole point and everyone can see the date on screen. It applies to the artifact that gets filed. So the question to settle before choosing a mechanism is which of the two you are making, and if the answer is a document of record, freeze it.

The pattern that works

Export the chart as an image at the moment the pack is assembled, and put the as-of date inside the image rather than only in the caption, so it survives being screenshotted, pasted into a summary email or printed. Restate the one number the slide exists for in the surrounding text, because a reader skimming a PDF on a phone reads words before they read axes. Then give one link to the interactive version, pointing at a page that holds that chart and nothing else. This is the same discipline that a chart in an email and a chart in a LinkedIn newsletter both force on you, for different reasons.

Size it for the room as well as the file. A slide read from the back of a meeting room and a slide read on a laptop have different legibility floors, and the usual failure is an axis label carried over from a web chart at a font size that works at full width and vanishes at slide scale.

Presenting from the live version without embedding it

If the plan is to drill into the chart when a question comes up, do it from a browser and leave the deck alone. A slide carrying the frozen image, the as-of date and a link works in the room, in the emailed PDF and in the minutes. Switching to the live chart for thirty seconds costs nothing and avoids every dependency above: no license, no tenant setting, no add-in, no blank frame in front of a board.

Building it in Quadesto

The practical problem with the frozen-image pattern is maintaining two versions of the same chart. Quadesto is built so that is one artifact: the chart is defined once, the interactive version is what you link to, and the static export for the slide is generated from the same definition with the as-of date stamped on it rather than typed into a caption by hand.

[QUADESTO-EMBED: a quarterly performance chart shown twice, once as the live interactive version and once as the slide-sized static export generated from it, with the as-of date rendered inside the image]

That removes the failure this whole page is about, which is a deck and a live chart drifting apart between the day the pack was written and the day somebody reads it back. Start on the free tier and try it on one board chart before the next quarter end.