A decade ago, most weather apps were free or cost a dollar or two. Today, the most popular ones charge $30-50 per year. CARROT Weather's premium tier runs $39.99 annually. Weather Underground gutted its free features after The Weather Channel acquired it. Dark Sky, beloved for its hyperlocal minute-by-minute forecasts, was bought by Apple and shut down entirely. On Reddit, the frustration is constant: why does it cost so much to check if it's going to rain, when the underlying data comes from government agencies that taxpayers already fund?
It's a fair question. But the answer is more complicated than "greedy developers." Here's where the money actually goes.
Most weather forecasts originate from a handful of government-run meteorological organisations. In the United States, the National Weather Service (NWS) operates under NOAA and provides forecasts, radar data, severe weather alerts, and historical climate records. All of this is public domain -- US taxpayers fund it, and anyone can access it for free via the NWS API.
In the UK, the Met Office provides similar data, though it operates differently. While basic forecasts are freely available, the Met Office also sells commercial data products and detailed model outputs. The European Centre for Medium-Range Weather Forecasts (ECMWF), widely considered the gold standard for medium-range forecasting, has historically charged for its high-resolution model data, though it opened up some datasets in recent years.
Other major sources include Environment and Climate Change Canada, the Japan Meteorological Agency, and the Australian Bureau of Meteorology. Each country's approach to data access and pricing differs, but the raw observational data -- temperature readings, barometric pressure, wind speed from weather stations -- is generally available from public sources.
So if governments produce the data and much of it is free, what are weather apps actually paying for?
Raw weather data from government agencies is designed for meteorologists, researchers, and other government systems. It arrives in specialised formats like GRIB2 (for model output) and CAP (for alerts). It's not formatted for a phone screen. Turning NWS GRIB2 files into the JSON response that shows "14 degrees and partly cloudy" on your iPhone requires significant infrastructure.
This is where weather API providers come in. Companies like Tomorrow.io, Weatherbit, Visual Crossing, and Open-Meteo take raw data from multiple government sources, run it through post-processing models, combine it with private data networks (personal weather stations, IoT sensors, satellite imagery), and serve it through developer-friendly APIs.
These APIs charge per request. A typical pricing structure might look like:
A weather app with 100,000 active users, each checking the weather a few times per day, might make 300,000-500,000 API calls daily. At commercial API rates, this can cost thousands of pounds per month. And usage scales linearly -- every new user adds to the bill.
After acquiring Dark Sky in 2020 and shutting it down in 2023, Apple integrated its technology into the built-in Weather app. Apple Weather now offers minute-by-minute precipitation forecasts, severe weather alerts, air quality data, and next-hour precipitation maps. It's free for every iPhone user.
But Apple Weather isn't really free. Apple funds it as part of the iOS ecosystem -- it exists to make iPhones more useful, which sells hardware. Apple can absorb the data and infrastructure costs because it makes its money from device sales, services subscriptions, and the App Store. The weather app is a feature, not a product.
This creates an awkward dynamic for independent weather developers. They're competing against an app that's pre-installed on every iPhone, costs the user nothing, and is backed by a company with $383 billion in annual revenue. The indie developer, meanwhile, is paying for every API call out of pocket.
Given that Apple Weather is genuinely good in 2026, why do people still pay for third-party weather apps? The answer lies in what Apple Weather doesn't do well.
Widgets and customisation. Apple Weather's widgets are functional but limited in design. Apps like CARROT Weather and others offer deeply customisable widgets -- different layouts, data density, complication styles, colour themes. For people who check the weather frequently, a well-designed home screen widget is worth paying for.
Multiple data sources. Apple Weather uses a single data pipeline (derived from the old Dark Sky models plus other sources). Some third-party apps let you compare forecasts from different models -- the GFS, ECMWF, UKMO, and others -- side by side. For weather enthusiasts and anyone planning outdoor activities where accuracy matters, seeing model consensus is valuable.
Radar and maps. Apple Weather shows precipitation maps, but dedicated weather apps often provide more detailed radar imagery, including future radar projections, storm tracking, lightning strike data, and multi-layered map views showing pressure systems, wind patterns, and fronts.
Personality and experience. CARROT Weather built its user base partly on its sarcastic, dark-humour commentary. As one indie developer put it on Reddit, "I wanted something more alive than the usual weather app." Some people pay for the experience, not just the data.
Understanding why weather apps cost money is one thing. Understanding why so many switched to subscriptions is another.
The one-time purchase model worked when weather data was cheap and API providers were generous with free tiers. An indie developer could charge $2.99 for a weather app, use a free or low-cost API, and the revenue from sales would cover server costs for years. But as the major API providers consolidated and raised prices, that model broke.
The economics are straightforward. A one-time purchase generates revenue once per user. But the API costs for that user continue for as long as they use the app -- potentially years. If an app charges $4.99 once and the API costs $0.02 per user per month, the developer has roughly 20 months before they're losing money on that user. With Apple taking 30% of the purchase price, it's closer to 14 months.
Subscriptions solve this by aligning revenue with costs. If the API bill goes up, the recurring revenue covers it. This is why nearly every weather app that once cost a few pounds now charges monthly or annually.
But "solves the developer's problem" and "feels fair to the user" are different things. When you're paying $3.33 per month for CARROT Weather Premium (the $39.99/year plan), you're paying more per year than the app used to cost as a one-time purchase. Over three years, that's $120 for a weather app. Many users understandably feel that's excessive for data that ultimately originates from publicly funded agencies.
Not every weather app has adopted the subscription model. Several approaches exist for developers who want to avoid recurring charges:
Using Apple's own WeatherKit. Apple offers WeatherKit, its weather data API, to developers. It includes 500,000 API calls per month for free (with an Apple Developer Program membership), which is enough for a small-to-medium app. Beyond that, it's $0.50 per 1,000 calls. An app using WeatherKit can keep its API costs low enough to sustain a one-time purchase model, at least at moderate user counts.
Open-source weather APIs. Open-Meteo is a free, open-source weather API that draws from public data sources including the DWD (Germany's national weather service), NOAA, and Environment Canada. It provides global coverage with no API key required for non-commercial use and reasonable pricing for commercial apps. This dramatically reduces the per-user cost.
Caching and efficiency. Smart caching strategies can reduce API calls significantly. Weather data doesn't change every second -- a forecast that's refreshed every 15-30 minutes is perfectly adequate for most users. An app that caches aggressively and avoids redundant requests can serve more users on fewer API calls.
One-time purchase with optional tips. Some apps charge once for the full feature set and offer optional tip jars or "buy the developer a coffee" mechanisms. This covers the majority of costs through the purchase price while allowing satisfied users to contribute more.
It's also worth considering the cost of free weather apps that aren't Apple Weather. The Weather Channel app (owned by Allen Media Group, which bought it from IBM) is free to download but supported by aggressive advertising. The app's privacy nutrition label on the App Store is extensive -- it collects location data, identifiers, usage data, and diagnostics, much of which is used for third-party advertising.
AccuWeather similarly monetises through ads and data collection. Its privacy label lists browsing history, location, identifiers, and usage data as collected and linked to your identity.
For these apps, you're not the customer -- you're the product. Your location data, checked multiple times daily, is extraordinarily valuable to advertisers. A free weather app that knows where you are, when you're there, and how often you check (correlating with whether you're planning to go outside) provides rich behavioural signals.
In this context, paying for a weather app is partly paying for the absence of surveillance. The $40/year for CARROT Weather Premium isn't just buying sarcastic commentary and custom widgets -- it's also buying a business model that doesn't depend on selling your location history.
Here's a practical framework for deciding what to spend on weather:
Weather apps cost more than they used to because the infrastructure between "government weather station records a temperature reading" and "your phone shows a clean forecast" is expensive. API providers, post-processing, server costs, and Apple's 30% cut all add up. Subscriptions aren't inherently greedy -- for many developers, they're the only way to keep the lights on when every user generates ongoing costs.
But that doesn't mean you have to pay $40 or more per year. The market in 2026 offers a genuine spectrum: Apple Weather for free, one-time purchase apps that use efficient data sourcing to avoid subscriptions, and premium subscription apps for enthusiasts who want every feature. The right choice depends on how much you care about widgets, data sources, and privacy -- not on whether you "should" pay for weather data that starts out as public information.
The weather itself is free. The forecast is the product. And like most products, you have more options than the most expensive shelf suggests.