The Mood Swing Index

Out of the 12,187 apps in this study, only 53 show an improving review trajectory, just 0.4% of the entire base. Improvement at that rate looks less like a category trend and more like noise. Compare that to 5,043 apps, or 41.4% of the base, that are declining, and you get an app economy where review momentum decays far more often than it builds. A few weak categories aren't dragging down the average here. Stable is the ceiling for almost every app, and once an app stops getting maintained or promoted, decline is what follows. If you're benchmarking a single app, the takeaway is simple: improving reviews put you in a group of 53 out of 12,187, and that alone makes you unusual.

Slice the data by category and News comes out worst, with 58.1% of News apps showing declining share. Utilities isn't far behind at 57%, then Social Networking at 55.9%, Travel at 55.7%, and Business at 55.2%. Each of those figures only describes that category's own population, not the full 12,187 study base, but the same pattern keeps showing up: apps built around frequent, disposable, or algorithm-dependent use, think news feeds, utility toggles, social feeds, tend to decline more often than apps people open once and then wander away from. Food & Drink (52.3%), Lifestyle (51.4%), and Photo & Video (50.3%) land in roughly the same range, all past the halfway mark within their own categories. Even Productivity, the lowest of the bunch at 44.5%, still sits close to a coin flip. One plausible read on this: apps that depend on constant content refreshes or on staying in a platform's algorithmic good graces, like News and Social Networking, run into volatility that slower utility categories rarely have to deal with. Users usually gripe about staleness in a review long before they bother deleting the app.

Developer Tools tops every other category for improving share, coming in at 2.9%, nearly double what Sports puts up at 1.7%, and it edges past Navigation's 1.5% by a fair margin too. Books and Reference tie at 1.3%, Finance follows at 1%, and News, Social Networking, Business, and Utilities all cluster close together, somewhere between 0.8% and 0.9%. Worth remembering here: these are shares within each category, not raw counts against the 12,187 base, and since only 53 apps improved across the whole study, even the top categories are really just talking about a handful of apps each. The interesting part is that News and Social Networking, despite also carrying some of the worst declining shares, still manage to squeeze out a small slice of improving apps. Decline isn't uniform within a category, so while most News apps are sliding, a few climb anyway, probably the ones that actually rethought their content strategy or pricing instead of just coasting.

Improving apps average $4.1, stable apps average $1.1, and declining apps average $0.5. That's not a small gap. It works out to roughly an eightfold difference between the top and bottom of the trajectory ladder. The pattern makes sense once you think about it: apps priced higher upfront usually come from teams with more on the line, which gives them a clearer reason to keep supporting the app long after launch. Free or steeply discounted apps, on the other hand, tend to get left behind once that first burst of downloads fades.

This runs against a pretty common assumption. Being cheap doesn't buy an app any extra goodwill or review momentum just because more people can access it. If anything, low-priced apps are the ones most likely to be sliding downward. If you only had one number to judge category health by, this would be it: price and trajectory move together here, and the free-to-declining pipeline holds up.

Stable apps sit at 4.7 on average, improving apps come in at 4.6, and declining apps land at 4.5. That's a gap of just 0.2 stars across the entire trajectory range. Star rating alone won't tell you whether an app is on its way down or on its way up. Most buyers look at that number first, yet it's actually the weakest signal for figuring out where an app is headed. Stable apps even edge out improving apps on raw rating, which seems backwards until you consider that ratings are lagging averages built up over months or years, while trajectory reflects what's happening right now. An app can hold onto a solid historical rating even while its recent reviews are sliding, and that 0.2 star gap is proof of how much decline a rating can mask while it's actually happening.

Average sentiment score runs from 0.9 for improving apps down to 0.5 for stable apps and -0.3 for declining apps, a spread of 1.2 points from top to bottom. Star ratings barely move across those same groups, shifting only 0.2 stars, so sentiment is catching something the star average simply misses. Declining apps stand alone as the only group with a negative average sentiment score. Reviewers there aren't just cooling off, their words are turning openly hostile, and the star average won't catch up to that reality for a while. It's the sharpest number to come out of this study: sentiment score beats star rating as an early warning signal by a wide margin. Anyone building a monitoring dashboard off this data should weight sentiment far more heavily than rating when trying to figure out where an app is headed.

Looking at the 12,187 apps in this study, 58.2% hold steady, 41.4% are slipping, and only 0.4% are actually getting better. Stable and declining sit just 17 percentage points apart, which is closer than you'd expect from an ecosystem everyone likes to describe as growth obsessed. For every three apps that manage to hold their ground, about two are bleeding review sentiment. Improving apps are so rare, at 0.4%, that the whole picture basically boils down to two camps: stable or declining. So any category level narrative about "improving apps" is mostly just talking about a thin sliver perched on top of that two state reality.