Building a Podcast Advertising Strategy From Scratch

Podcast listenership had grown steadily since Serial proved in 2014 that a podcast could pull a mainstream audience, and by 2018 dynamic ad insertion technology had matured enough that podcast advertising finally worked like a real, manageable media channel rather than a one-off favor from a host you happened to know. Here’s how to build a first campaign from nothing.

Step one: pick shows by audience fit, not just size

Download numbers get all the attention, but a smaller show with a tightly matched, highly engaged audience routinely outperformed a much larger show with a loosely related one. Start by listing what your customer sounds like, then look for shows whose audience matches that, rather than starting with a ranked list of the biggest shows in a broad category.

Step two: understand host-read versus produced ads

Podcast advertising in 2018 split between host-read ads, where the show’s host delivers the pitch personally, often with genuine or performed enthusiasm, and pre-produced spots inserted the same way a radio ad would be. Host-read ads generally converted better because listeners trusted the host’s voice, but they cost more and required giving up some control over exact messaging. Pre-produced spots gave more message control at a lower cost but weaker trust transfer. Neither is automatically right; the choice should match how much you’re willing to trade control for credibility.

Step three: use dynamic ad insertion where it’s available

By 2018, platforms like Megaphone and Art19 had built dynamic ad insertion tools letting publishers swap ads into an episode after it was already published, rather than baking a fixed ad permanently into the audio file. This meant a campaign could run across an entire back catalog of episodes, not just new releases, and let advertisers adjust or end a campaign without waiting for old episodes to age out of relevance.

Step four: build attribution in before launch, not after

Podcast ads don’t offer a click. Set up a unique promo code or a dedicated landing page URL per show before the campaign launches, not after someone asks why you can’t explain the results. This is the single most commonly skipped step, and skipping it is why so many first-time podcast advertisers ended up unable to say whether the campaign worked at all.

Step five: plan for frequency, not a single flight

A single ad read rarely moved the needle on its own. The campaigns that performed ran a sustained flight, often four to eight weeks of consistent placement on the same shows, giving the audience repeated exposure rather than one shot to be remembered. Budget for that sustained run from the start rather than testing with a single week and judging the whole channel on it.

Step six: expect a longer sales cycle in the data

Podcast listeners often act on an ad days or weeks after hearing it, not in the next ten minutes, since most people are driving, working out or doing something else entirely while listening. Build reporting windows that account for that delay rather than judging performance against the tighter timelines you’d expect from search or display.

Putting it together

A business starting from zero in 2018 would typically shortlist two or three well-matched shows, choose host-read ads if budget allowed, set up unique promo codes per show, commit to a multi-week flight rather than a single test read and give the results a few weeks past the campaign’s end before drawing conclusions. That’s a modest, disciplined setup, and it’s enough to get an honest read on whether podcast advertising belongs in the ongoing media mix.

---

More from Paid Advertising, the 2018 archive or the Digital Business Marketing Awards.