Does Empire-Building Hurt Quality?

Average ratings don't fall off a cliff, they sag like a valley: 4.3 for solo publishers, 4.1 for those with 2-5 apps, 4.0 for 6-20 apps, dropping to 3.9 at 21-100 apps, then climbing back to 4.0 at 100+ apps. Publishers in that 21-100 range are stuck in an awkward middle: too big to give each app personal attention, but still lacking the systems that would make scale actually manageable. The bounce-back at 100+ apps isn't proof that size helps. Only 139 publishers ever reach that scale, so the 4.0 average there just reflects who stuck around long enough to get there, not any built-in advantage. Quality control falls apart in the 21-100 bucket, while the 100+ bucket is really a survivors' sample, publishers who either figured things out or caught a break. So if you're sizing up a publisher's output, a large catalog in the 21-100 range should read as a warning sign, not a credential.

Of the 357,961 publishers in this study, 260,145, or 72.7%, publish exactly one app. Another 61,274 (17.1%) run 2-5 apps, and the numbers keep collapsing from there: 29,037 (8.1%) at 6-20 apps, 7,366 (2.1%) at 21-100 apps, and just 139 publishers, less than 0.1% of the total, operating more than 100 apps. This matters because the 'does empire-building hurt quality' question is really a question about a tiny fringe of the market. The vast majority of the quality story in this industry is a solo-developer story, and the empire-building effect, whatever it is, only ever touches a sliver of publishers. Anyone generalizing about 'app factories ruining quality' is talking about 2.1% of publishers at most, and about an almost invisible 139 at the extreme end.

Sort portfolios by ratings per app and the order gets strange pretty quickly: publishers with 2-5 apps top the chart at 5,232.5, publishers with 100+ apps follow at 3,930.2, the 6-20 app group comes in at 2,724.4, 21-100 apps sits lower at 2,226.1, and solo publishers bring up the rear at 2,090.1. Solo publishers actually post the highest average rating, 4.3, yet they collect the fewest ratings per app of anyone. So quality and reach clearly don't move together at that end of things. The 2-5 app group looks like it has hit a sweet spot, just enough apps to cross-promote and build a following, but not so many that attention gets spread thin. The 21-100 app bucket is the weak link across the board, which lines up with what the ratings already suggested: this group struggles on both quality and engagement. Then the 100+ app bucket jumps back up to 3,930.2 ratings per app, and that smells like survivorship bias to me. Publishers who grow a catalog that big have usually figured out a formula that works, even if most publishers never get anywhere near that scale.

The 100+ app bucket averages 4.0, but that number falls apart once you look at individual publishers. ChowNow sits at 4.9 and satoshi yoshida at 5, both near the top of any rating scale, with Exam Prep OU (4.7), YAUHEN BAHDANOVICH (4.6), and Townsquare Media, LLC (4.6) close behind. At the other end, Razvan IVAN sits at 2 and CINC Systems at 2.6, rough scores for a consumer app by any measure. Magzter Inc. (3.3) and fly sky pty ltd (3.1) land somewhere in the middle. The bucket average of 4.0 is a statistical convenience, and it doesn't describe how any single giant is doing. Running 100+ apps doesn't predict your rating. It means you've stuck around long enough to get counted. If you're using portfolio size as a stand in for quality, this is about as clear a counterexample as you'll find.

Publishers who stick to one app average a 4.35 rating, while those running multiple apps settle at 4.06, which works out to multi-app quality sitting at 93.3% of what the single-app crowd manages. The wider bucket data holds up this pattern: quality slides as portfolios grow, so this isn't some rounding fluke, it's a real gap. Still, the drop is modest rather than a collapse. Add more apps to your lineup and your rating takes a hit, but it doesn't fall apart. Worth looking at the publisher counts behind this split too. There are 260,145 single-app publishers against 97,816 multi-app publishers, meaning the multi-app group comes in at 37.6% the size of the single-app group. Roughly a quarter of the publishers in this study have spread themselves across several apps, and on average they're paying a small but measurable quality price for it. A second or third app will cost you something, sure, but nowhere close to what you'd pay running 21-100 apps.

Here's a number that undercuts the entire "app factory" story: the 139 mega-publishers in this study, each running more than 100 apps, account for a combined 0.03% of all ratings. People tend to talk about these publishers like they're giants, which makes that figure land as a genuine surprise and forces a rethink of the whole empire-building angle. These companies simply don't dominate the ratings pool. They barely show up in it. Sure, each one has its own story worth digging into (the ChowNow versus Razvan IVAN split is a good example), but added together they hardly register against the total volume of consumer feedback out there. So any quality narrative built around the 100+ app bucket, rebound or not, only describes a sliver of real ratings activity, not the place where most App Store engagement actually lives. That mass sits with the much larger single-app and 2-5 app crowd. If you're hunting for where App Store quality and attention really concentrate, look past the mega-factories.