How do I track podcast ads?

Podcast ads are heard, not clicked, so you usually track them with things a listener can remember. Give each show a short address that redirects to a tagged page, plus its own discount code. Ask buyers where they heard about you. People who just type your name slip past all three, so treat the count as a floor.

A podcast ad has no button. The host says your name, maybe a code and maybe an address, while the listener drives, runs or does the dishes. Whatever they do next happens later and somewhere else.

So the usual answer hands the listener something to carry. The short address is for people who remember it. It redirects to a page whose link carries UTM tags, so GA4 can file the visit. Set utm_medium to audio and the visit usually lands in GA4's Audio channel. Google describes Audio as visits from ads on audio platforms, such as podcast platforms.

The code is for people who remember a word instead of a web address. Shopify's Sales by discount codes report groups those sales under the code's name. The checkout question catches people who remember the show but neither of the other two.

All three are worth setting up. None of them sees the listener who typed your store's name a week later, and that listener is the hard part.

What one store's data shows

One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.

Start with Direct. On the Channels sheet it holds 57.7% of revenue, and the share is identical in last click, first click and touched views. GA4 files a visit as Direct when someone enters your address straight into the browser. A listener who heard your name on Tuesday and typed it on Friday arrives exactly like that.

The first click column matters for podcasts. The episode happens before the first visit, away from your site, so no first-click view can hand the show any credit. For 57.7% of revenue on the Channels sheet, the first touch the export can see is already Direct.

GA4 does try to carry tags forward. Its help says sessions that start from a direct entrance are attributed to the UTM values it already holds for that user. So a typed visit usually keeps a podcast tag only if the same browser came through your tagged address earlier.

On the Journeys sheet, journeys with 1 touch hold 79.5% of revenue, and those buyers took 0.5 days to buy. One visit, same day, done.

That is what a convinced listener looks like in GA4: heard the pitch, typed the name, bought. It is also what a loyal customer reordering looks like. The export cannot tell the two apart, because it records visits, not what anyone heard.

Journeys with 2 to 3 touches hold 12.2% of revenue on the Journeys sheet and took 12.5 days to buy. A listener who wanted to think it over could sit in this row, between a search, an email and a typed visit. Nothing in the row says a podcast started it.

What the export cannot show is how many of those journeys an episode started. It cannot even say whether this store ever bought a podcast ad.

Why do codes and short addresses miss sales?

Each one asks the listener to remember something exact at the right moment. A listener can easily remember your brand and forget the rest.

The short address needs word-for-word recall. Anyone who types your homepage instead becomes Direct. The redirect can also lose its tags on the way. GA4's help lists redirects among the things that strip UTM parameters. Shopify warns that URLs with query strings might not work as expected with redirects.

The code needs a reason to be typed. If your store runs a bigger sitewide sale that week, the listener uses that discount, and the show gets nothing. Codes also leak: once one reaches a coupon site, people who never heard the episode use it.

The checkout question needs a good memory. People forget where they heard things, and memory rounds off. Read the answers as a rough split between shows, not a count.

Nothing catches the searcher. A listener who searches your brand next week arrives as Organic Search, or as Paid Search if your brand ad wins the click. Then your brand campaign collects the credit, and the show looks like it did nothing.

What can podcast tracking not tell you?

Whether the show caused the sale. A code records who typed it, and an address records who used it. Neither says whether that buyer was coming anyway.

The closest you get without a test is a before-and-after. Mark each air date, then set the week after it against the same weekdays the week before. GA4's date picker has a Compare option called Previous period (match day of week) for exactly this. Watch Direct, Organic Search and new customers, not only the code.

If only the code moves, the code tells most of the story. If Direct and brand searches jump too, your tags and codes undercount. If nothing moves, one episode may simply be too small to see, so judge a run of episodes, not one.

Over a longer stretch, a mix model can read podcast spend against weekly sales without any tags at all. Meta's Robyn is one free example, if you have the history and an analyst.

What to do this week

1. Tag the landing link with utm_medium=audio. Use utm_source=podcast, utm_medium=audio and the show's name in utm_campaign. Pass: a test visit shows up in GA4 under Reports > Acquisition > Traffic acquisition, in the Audio row. Fail: it sits in Unassigned, the bucket GA4 uses when no channel rule matches.
2. Type the short address like a listener would. Open a private browser window and type it exactly as the host will say it. Pass: the page loads and the address bar still shows the utm tags. Fail: the tags are gone, so give the show a page of its own and count visits to that page instead.
3. Compare the week after an episode. In Traffic acquisition, pick the seven days after the air date, click Compare and choose Previous period (match day of week). Pass: Direct or Organic Search rises against the week before. Fail: both stay flat, so the episode did little or its effect is too small to see in one week.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: Default channel group (Google); Understand (direct) / (none) traffic (Google); Scopes of traffic-source dimensions (Google); Traffic acquisition report (Google); Change and compare date ranges in reports (Google); Sales reports (Shopify); Creating and managing URL redirects (Shopify)

Related answers

• All UTM & tracking questions
• How to track podcast ads in GA4 and Shopify, step by step
• How do I track podcast ads for a fashion brand?
• Podcast attribution: measuring what no cookie can track
• Direct traffic, defined