The Long Way Round
Sponsor

Host, The Long Way Round · Weekly since 2021 · 214 episodes · Two host-read slots per episode · booked a month out

A download is not a listen.

A weekly solo show about small infrastructure and the people who maintain it. Every number on this page states the window it was measured over and the standard it was measured against, because a media kit that does not is asking you to guess.

The numbers, defined Rates

Every figure, with its definition

18,400 downloads. 9,700 past the mid-roll.

The second figure is 53% of the first, and it is the one a mid-roll is actually sold against.

Yellow figures are downloads. Blue figures are measured playback.
Figure Value What it actually counts
Downloads, 7 days IAB v2.2 certified · per episode, median of last 12 18,400 The number every media kit leads with. It counts a device that fetched at least a minute of the file, once, deduplicated over 24 hours. It does not count a person.
Downloads, 30 days IAB v2.2 · same episodes 24,900 The same episodes given three more weeks. If a kit quotes only this figure, it is quoting a bigger number for the same show — which is fine, as long as it says so.
Unique devices, 30 days Deduplicated by device, not by person 14,200 Closer to an audience size, still not one. A phone and a car head unit are two devices and frequently one listener.
Median completion Of episodes started, across last 12 71% From platform analytics rather than downloads. Half of everybody who starts an episode is still there at 71% of its length.
Past the mid-roll Reached 00:14:00, median episode 9,700 The number a sponsor should actually price against, and the one nobody publishes. It is roughly half the seven-day download figure, which is the whole point of this page.
Past the post-roll Reached final 3 minutes 6,100 Published because a post-roll slot costs less for a reason, and pretending otherwise sells a sponsor something they did not get.
Same show, three honest numbers, and each one is smaller than the one above it. A sponsor is buying the second bar.

The gap between the first row and the fifth is the whole argument. Both are honest numbers about the same show; one is what a file server saw and one is what a person did. Any media kit can quote the first. A kit that quotes the fifth is telling you what it is actually selling, and I would rather price against a smaller true number than a larger vague one.

Last 10 episodes · 30-day figures

The top one did 2.4× the median.

Median 25,300. That spread is normal and it is invisible in an averaged media kit.

Sorted by downloads, not by date — the spread is the point
No. Episode Downloads Completion
#214 The last switchboard in the county 61,200 78% completion 78% completed
#211 The man who reads the river gauge 31,800 81% completion 81% completed
#207 Rural broadband, four years later 28,700 67% completion 67% completed
#213 Who owns the water tower 26,400 74% completion 74% completed
#209 A grain elevator with a fax machine 25,300 70% completion 70% completed
#205 What a county road actually costs 24,100 71% completion 71% completed
#212 Two hundred miles of unlit road 23,100 69% completion 69% completed
#210 Snow fences, and why they are angled 22,600 72% completion 72% completed
#206 One traffic light, three jurisdictions 21,500 73% completion 73% completed
#208 The bridge inspection nobody attended 19,800 64% completion 64% completed

The top episode did about two and a half times the median and it is not the one I would have predicted. That spread is normal, it is invisible in an averaged media kit, and it is why a sponsor buying "an episode" is buying a distribution rather than an average. Sponsorships here are sold against the median and reported against the actual, which means some months you get more than you paid for.

Rates

CPM against the mid-roll, not the download.

CPM here is calculated against the mid-roll figure, not the download figure. Against downloads the same slot would look like a much lower CPM and would be selling attention the show does not have.

Six lines, including the two nobody publishes
Placement Price CPM Note
Pre-roll, 30 sec $180 $18 CPM Priced against 9,700. Heard by nearly everyone who starts, which is why it is not the cheapest slot despite being the shortest.
Mid-roll, 60 sec $340 $35 CPM The one worth buying. Host-read, in my own words, placed at a break in the episode rather than at a fixed timestamp.
Post-roll, 30 sec $110 $18 CPM Priced against 6,100, not 9,700. It costs less because fewer people hear it, and that is the only honest reason a slot should cost less.
Four-episode run $1,150 — Mid-rolls, about 15% off. Repetition is most of what makes host-read advertising work and one episode rarely does anything.
Make-good Automatic — If an episode lands more than 20% below the median at 30 days, the next one is free. Not negotiated — it is in the confirmation email.
Categories I decline Listed — Gambling, crypto trading, and anything making a health claim. Stated up front so nobody wastes a pitch.

“She sent a rate card with the mid-roll listener number on it, which was half the download number in the same email. I have bought from a hundred shows and it was the first time I could compare two of them honestly.”

Media buyer, regional advertiser — 2025

Open the Media Kit →

Common questions

Mostly about which number is which.

Why publish the smaller number?

Because it is the one being sold. A sponsor buying a mid-roll is buying the people who were still listening at fourteen minutes, and that is 9,700 rather than 18,400. Quoting the larger figure is not lying exactly — it is answering a different question than the one being asked.

What does IAB v2.2 mean?

An industry standard for counting a download: at least a minute of the file requested, deduplicated per device per 24 hours, bots excluded. It is the difference between a comparable number and a number. If a show cannot tell you which standard it uses, it does not have one.

Downloads or listeners?

Downloads are what a file server saw. Listeners are what platform analytics infer, imperfectly, from playback. This page publishes both and labels which is which — the blue figures are measured playback, the yellow ones are downloads.

Why is the top episode so far above the rest?

One episode got picked up by a much larger show and did two and a half times the median. That is the normal shape of a back catalogue and it is exactly what an average hides. You are buying the median; some months you get more.

Do you take pre-recorded ads?

Only if the alternative is not working. Host-read outperforms programmatic on this show by a wide margin, and I would rather turn down a booking than run a spot that will not perform and then be blamed for the CPM.

What reporting do I get?

A one-page report fourteen and thirty days after the episode: downloads, completion, mid-roll reach, and the make-good calculation if it applies. Sent without being asked, because a sponsor chasing numbers is a sponsor who will not renew.