Good morning,

I have to say, it is hard enough to write about either Shopify or Spotify without using the wrong name; I’m pushing my luck writing about both in the same post!

On to the update.

Shopify’s Evolution Follow-up

First off, I made a mistake in Shopify’s Evolution when I said that Shop Pay was the same thing as Shopify Payments; Shopify Payments is the merchant-facing solution which can be used instead of a merchant account. Shop Pay is the consumer-facing product that allows for one-click check-out. Interestingly, Shop Pay can be used with merchant accounts. I regret the error, although the broader point about Shopify integrating into payments (which drives an increasing amount of the company’s revenue) does hold.

Secondly, an investor writes, “Fulfillment centers don’t get the same multiple that software-based platforms do. Major mistake for SHOP – lost their investor base.” This certainly makes sense, but the degree to which it is correct speaks to the challenges of being a public company and why it often requires a founder to make hard decisions. The reality is that the world in which it made sense to be a software platform floating above the fray, leaving advertising to Facebook and fulfillment to disparate 3PL companies, is one that is going away; that was why I started with the anecdote about cheap ads on Google, to demonstrate how much the field of play has shifted. Shopify could stay lightweight and perhaps keep their multiple, but at the cost of significantly increased strategic risk. As I note below the future is about larger businesses, not smaller ones, and Shopify has no choice but to adjust.

Third, to continue from the last point, the fact of the matter is that Apple has fundamentally changed the rules of the game in a very profound way, making it increasingly impossible to be a small scale player. I wrote about this letter from Apple’s Senior Director of Global Privacy Jane Horvath in 2020’s Privacy Labels and Lookalike Audiences, but it’s worth revisiting:

Advertising that respects privacy is not only possible, it was the standard until the growth of the Internet. Some companies that would prefer ATT is never implemented have said that this policy uniquely burdens small businesses by restricting advertising options, but in fact, the current data arms race primarily benefits big businesses with big data sets. Privacy-focused ad networks were the universal standard in advertising before the practice of unfettered data collection began over the last decade or so. Our hope is that increasing user demands for privacy and security, as well as changes like ATT, will make these privacy-forward advertising standards robust once more.

This is a profound misrepresentation of how the Facebook ecosystem, which includes Shopify, worked. Yes, Facebook had a massive data set, but that was because no one else had any data — they gave it all to Facebook because it was better for everyone to work as a collective than to try and build some piddling ad business on their own (we’ve seen how well that worked for newspapers). Destroying that ecosystem doesn’t suddenly empower all of these small businesses that relied on Facebook advertising; it just makes them unviable.

Fourth, and to that end, the reason why Shopify needs to load up on data collection and management for the purpose of advertising is precisely because they are the only entity positioned to do so. Very few Shopify merchants — probably none of them — have the scale to effectively manage advertising on their own; Shopify needs to form a new collective because in the Apple-defined Internet small players are road kill for whoever can get to scale first.

Fifth, to expound on the advertising network concept, it’s very important to note that the best advertising entails knowing who saw an ad and who converted; to date that has meant owning consumer-facing inventory like the Facebook feed (so you know who sees an ad) while having SDKs embedded at the point of conversion (so you know who converted). ATT breaks that SDK link, which means that whatever comes next is going to be worse. An ad network on the consumer-facing side is going to be blind to conversions; an ad network on the conversion side that buys ads at scale is going to be blind as to who actually saw an ad. Both sides are going to have to try and fill in the missing pieces probabilistically, which is going to be fundamentally less accurate than the previous deterministic approach.

That means that the question for a theoretical Shopify Advertising Services that acts as a first-party for all Shopify merchants, collecting all of their data and buying ads on their behalf, is not whether it can be more effective than the pre-ATT Facebook-centric approach; that is impossible. It’s whether or not it can be better than whatever other alternative there might be, including Facebook’s own solutions. Even that, though, is not quite right: Facebook’s probabilistic-based new system will work better the larger the scale of the advertiser; this both makes intuitive sense and has been noted multiple times by Facebook executives. That in and of itself is a reason for Shopify to become the largest advertiser of all, making Shopify the absolute best place to achieve effective Facebook advertising in a post-ATT world.

Leaving aside all of these details, though, I want to emphasize the broader trend that undergirded not just this article, but also Digital Advertising in 2022 and Gaming the Smiling Curve: the seemingly wide-open terrain of the Internet and the economic promise of a new economy filled with niche businesses made possible by the entire world being their addressable market is fading. It’s hard to see how targeted businesses work if targeted advertising is banned; what is striking is how that ban came about, not via regulation, but via the largest company in human history, cheered every step of the way.

Spotify Acquires Chartable and Podsights

From The Verge:

Spotify is making more podcast acquisitions. The company announced today it’s acquiring both Chartable and Podsights — two of the most prominent podcast marketing and ad attribution companies. The deal price hasn’t been disclosed, but this marks the first major acquisition the company has made this year in a long line of audio purchases.

Both Podsights and Chartable allow podcasters and networks to include tags in their shows that are used to track who listened, if they heard an ad, and whether they took action upon hearing it. Spotify says it plans to use Podsights’ technology outside podcasting and will bring it to the “full scope of the Spotify platform, including audio ads within music, video ads, and display ads.” The Chartable acquisition appears to be more directed toward podcasters themselves rather than advertisers, particularly because of its technology like SmartLinks.

The recent discussion of digital advertising provides excellent context for understanding these acquisitions. As I just noted, effective digital advertising depends on knowing who saw your ad and who converted; this is a much easier problem to solve when everything is digital, which is the first problem when it comes to podcast advertising, where the ad is analog — sound in your ears. Right off the bat that means that any effective measurement system is going to be probabilistic in nature, not deterministic.

What a service like Podsights does is try and catch signal on both sides. First, Podsights integrates with a podcast host so that it can capture data like the IP address, user agent, device data, etc. of every downloaded podcast that includes an advertiser’s ad. Advertisers then include a Podsights pixel (like the Facebook pixel) into their website or landing page, so that Podsights can capture the same sort of information from whoever visits. Podsights then matches up the data, along with some fancy math to account for devices that switched IPs, etc., making it possible to report to the advertiser how effective a particular campaign was, without needing to rely on the user remembering special URLs, discount codes, etc.

A few points about this methodology:

• First, as noted, this is much less accurate then simply tracking a unique identifier like the IDFA or GAID, or using a cookie.
• Second, there is an argument that this approach is also less private, given that an IP address includes rough location information (on the flip side, IP addresses change frequently).
• Third, this approach is, in broad strokes, exactly what the broader advertising ecosystem is moving to post ATT.

This raises an additional point: the axe that has yet to fall is an expansion by Apple of Private Relay, which obfuscates a user’s IP address; right now it works with Safari traffic but not app traffic. Apple could very well extend it to all iPhone traffic, or, as Eric Seufert mused, to SDKs specifically. Private Relay as it is today, though, is already a risk factor for measured podcast advertising.

Chartable, meanwhile, offers a very similar product: it makes sense that Spotify will combine them into one service for greater scale; meanwhile, Chartable has done quite a bit of work creating analytics for podcasters, which Spotify is going to incorporate into Megaphone, its podcast hosting service for the open podcast ecosystem. Making Megaphone more attractive as a podcast host enhances Spotify’s three-tier podcasting strategy, which I wrote about when Megaphone was acquired:

This acquisition certainly seems to suggest that Spotify agrees with the importance and permanence of the open podcast ecosystem — and that I had the company’s ultimate ambitions right at the beginning. Ideally, podcast listeners will move to Spotify, where the company will be able to know far more about them then they will on any other platform, and be able to build a dynamic ad platform (and, per last week’s Daily Update, upgrade them to ad-free subscription plans).

However, for podcast listeners that want to stick with the open ecosystem, Spotify will be able to leverage their ad network and listener base into the superior ad network for podcasts broadly. Sure, Megaphone may never have CPMs that rival on-network Spotify listeners, but they will be much higher than alternative podcast advertising networks that lack Spotify’s ability to understand a portion of the market and probabilistically apply that understanding to everyone else.

Those are the top two tiers; the bottom tier is podcasters who host elsewhere, who can now be reached with Podsights. The goal is to make all three into an addressable market for advertisers working with Spotify; at the same time, the higher the tier, the better the knowledge of who the listener is, and the better the measurement, which means higher prices. That is why Spotify is still incentivized to move listening up the ladder through levers like exclusives.

What is worth noting, particularly for those angsty about Spotify’s moves in the space, is that the fact that someone is trying to increase podcast monetization is a good thing; sure, larger, long-established podcasts don’t want things to change, but the better the advertising the more money there is, and the more money there is in the ecosystem the more opportunities there are for creators.

---

This Update will be available as a podcast later today. To receive it in your podcast player, visit Stratechery.

The Stratechery Update is intended for a single recipient, but occasional forwarding is totally fine! If you would like to order multiple subscriptions for your team with a group discount (minimum 5), please contact me directly.

Thanks for being a subscriber, and have a great day!