Every winter, millions of skiers hit the slopes across North America, completing an estimated 78.4 million ski visits each season. Some 15.8 million ski visits—roughly 20% of the North American market—happen at a property owned by Vail Resorts, the premier mountain resort company in the world.
Vail Resorts runs the #1, #2, and #3 most-visited ski resorts in North America: Vail Mountain Resort, Breckenridge Ski Resort, and Park City Resort. In total, the company owns 42 mountain resorts, in the Rocky Mountains, Pacific Northwest, Northern California, Switzerland, Australia and other top destinations. The company’s Retail and Rental business unit is responsible for retail sales and gear rentals at its resorts.
Every year, the company’s Retail and Rental business unit runs targeted, omnichannel ad campaigns to existing and prospective guests, using a mix of email, direct mail, promotional programs, digital marketing, and traditional media.
During the most recent ski season, Vail Resorts’ Retail and Rental division worked with Start.io to build targeted ad campaigns using a mix of first-party guest data from Vail Resorts, and audience segmentation and targeting data from Start.io.
Start.io targeted likely skiers and snowboarders living in specific cities near Vail Resorts properties. The results were impressive, with Start.io beating the performance of competing ad-serving platforms in cost-per-click (CPC), clickthrough rate (CTR) and cost per mille (CPM).
Cost-per-click performance exceeded Vail Resorts’ Retail and Rental benchmarks by an average of 16%.
“We have been pleased with the results seen from Start.io for our Retail and Rental businesses, with improved performance across many of our key metrics when compared to other partners,” said Jon Secrett, Senior Manager of Vail Resorts’ Retail and Rental business unit. “The team is knowledgeable and responsive, helping to drive efficiency and greater performance.”
Download the case study here.
Written by Elan Carter
Elan Carter is the Vice President of Growth and Business Development at Start.io. He’s worked in the advertising industry for more than 15 years, and lives in New York.
POSSIBLE 2025 is now in the books—can you believe that this must-attend industry event is just 3 years old? It’s been a couple of weeks since we all got back from Miami Beach and compared notes. Here are my big takeaways from three very packed days:
My first takeaway upon arriving at the Fontainebleau on Monday morning was, ‘Did our industry grow 10x?’
The hotel’s lobby was so packed, you could barely breathe. It was like taking the F Train during rush hour; I couldn’t hear myself think. And yet, when I looked across the lobby, I saw a lot of people in the advertising industry who were so happy to reconnect with friends and colleagues.
Second takeaway: For an event that launched in 2023, there has been consistent buzz that POSSIBLE is on track to become the new Cannes Lions. The heat, the attire, the back-to-back-to-back happy hours, rosé or Aperol spritz—heck, even the overpriced cups of coffee reminded me of southern France. Yacht Row in Miami was less impressive, but who knows what next year will bring.
Third, in terms of actual rumblings, or my version of what I’m hearing, there was a lot of talk about what’s next for Google’s adtech business. What will they choose to (or be forced to) spin off or sell?
I also got into a lot of ‘How about them cookies?’ conversations this year.
Regardless of what’s happening at Google, there was continuous chatter around advertising attention, identity, and addressability. More companies want to expand into additional channels, diversify their revenue streams, and figure out what’s actually AI—and what says it’s AI, but really isn’t.
Finally, mobile-centric companies like Start.io tend to get identified as ‘performance’ or ‘outcome-driven’ companies—that’s a partial misconception that I talked about a lot with people at POSSIBLE.
It’s not that there’s anything wrong with performance or brand performance, but at the Start.io cabana, I found myself tirelessly explaining how our user segmentation and audience curation capabilities help our clients optimize their campaign results with high-quality, impactful targeting.
The quality play is not new, but it is a path for brands and technology partners to optimize ad strategies across the industry. Ultimately, it’s about publishers who work hard on the content we all consume being able to capitalize on their hard work.
What’s next? Father’s Day in Cannes—I’ll see you there!
For the second year in a row, marketers will spend more money on digital video ads than on traditional linear TV, according to a new report from the Interactive Advertising Bureau.
The IAB estimates that 58% of all video ad spend in 2025 will go to digital channels via CTV, online video, and social video. The remaining 42% will go to traditional ads on linear TV.
That’s a dramatic shift in ad budgets over the past 5 years. In 2020, 71% of all video ads ran on traditional linear TV, with just 29% going to digital video, according to the IAB.
One of the fastest-growing ad types
Digital video ad spend has more than doubled since 2020, when brands spent an estimated $26.2 billion on the format, according to the IAB. That number has steadily grown each year, rising to $63.8 billion in 2024.
In 2025, the IAB estimates that digital video ad spend will grow another 14%, to $72.4 billion.
This shift is partly due to linear TV’s shrinking subscriber base. It’s estimated that traditional TV services lost nearly 15 million subscribers in the U.S. in 2023 and 2024.
This shift is also due to the improved targeting capabilities found with digital video ads, which allow brands to target consumers based on time, location, and individual interests.
Consumer packaged goods continue to outspend every other advertising category on digital video ads, with roughly $1 out of every $5 going to CPG ads, according to the IAB.
U.S. estimated digital video ad spend by category (2024 vs. 2025 growth %):
- $14.3 billion: CPG brands (+13%)
- $8.4 billion: Retail brands (+18%)
- $7.3 billion: Technology brands (+5%)
- $6.3 billion: Pharmaceutical brands (+19%)
- $6.2 billion: Entertainment and media brands (+3%)
- $6.1 billion: Automotive brands (+15%)
- $5.1 billion: B2B brands (+22%)
- $5.0 billion: Restaurant brands (+17%)
- $5.0 billion: Financial brands (+6%)
- $2.4 billion: Travel brands (+18%)
- $1.7 billion: Wellness brands (+19%)
- $0.9 billion: Clothing brands (+8%)
Source: Digital Video 2025: Ad Market Size and Growth Projections, IAB
Marketers say CTV is a ‘must buy’
In 2025, 68% of marketers now consider CTV a ‘must buy’—i.e. necessary for their media plan, according to an IAB survey of more than 360 advertisers.
Linear TV now sits at the bottom of the priority list with marketers, with just 33% saying they consider local broadcast/cable TV a ‘must buy.’ Audience-indexed linear TV is a ‘must buy’ with just 27% of marketers, according to the IAB.
Advertisers are largely shifting existing ad budgets toward CTV. Asked where they’re getting the money to spend on CTV ad campaigns, marketers say they’re:
- 36%: Reallocating budgets from linear TV
- 36%: Reallocating budgets from social media
- 34%: Reallocating budgets from OLV (excluding YouTube)
- 33%: Reallocating budgets from other types of traditional ads
- 32%: Reallocating budgets from paid search
- 31%: Reallocating budgets from digital display ads
- 30%: Reallocating budgets from out-of-home campaigns
- 27%: Reallocating budgets from digital audio/podcasts
- 23%: Reallocating budgets from gaming campaigns
- 10%: Reallocating budgets from non-media marketing dollars
Caveat: Digital ad projections could change in 2025
In their report, the IAB notes that ongoing economic uncertainty could impact their 2025 projections, as advertisers potentially pull back on budgets due to pressure from tariffs, geopolitical conflict, and changing consumer sentiment.
In a separate and unrelated report, analysts at eMarketer estimate that linear TV ad spend could decline between $2.78 billion and $4 billion this year, largely fueled by global U.S. tariffs. Spending on CTV ads could end the year flat, or slightly up, according to eMarketer. In other news, Nielsen estimates that 72.4% of all TV viewing in Q1—across both linear TV and CTV—was ad-supported. The remaining 27.6% of TV viewing appeared on ad-free platforms.
An estimated 8 out of 10 developers who build a subscription-based mobile app fail to earn $1,000 per month in subscription revenue within their app’s first 24 months, according to new data from RevenueCat.
This finding suggests developers who want to reach revenue milestones should pursue a hybrid monetization strategy, offering a mix of subscriptions, in-app purchases, and advertising. Most mobile apps attract a diverse range of customers, some of whom will insist on an ad-supported option.
Building a successful, subscription-based mobile app isn’t easy. RevenueCat, which makes a tool that helps developers manage their subscriptions, estimates that fewer than 6% of mobile apps earn $10,000 or more per month in recurring revenue by year 2:
- 80.83% fail to earn $1,000 in monthly recurring subscription revenue (MRR) by year 2
- 19.17% earn $1,000+ in MRR by year 2
- 11.92% earn $2,500+ in MRR by year 2
- 8% earn $5,000+ in MRR by year 2
- 5.34% earn $10,000+ in MRR by year 2
Mobile consumers are fickle, no matter what type of app you’re building. It’s estimated that the average app loses 77% of their daily active users by day 3 after install.
1% customer conversion rate
Subscription-based apps can face an even steeper challenge, because they ask consumers to find enough value in their content that they’re willing to pay to access to the content. In a study of 75,000 subscription-based mobile apps, RevenueCat found that the median-performing app was able to convert just 1% of their downloads to a paying customer by day 35.
In practical terms, imagine that 10,000 people downloaded an app on day 1. By day 35, just 100 people would have converted to subscription-paying customers. That’s the median—half of subscription-based mobile apps fail to achieve a 1% conversion rate after 35 days.
Health and fitness apps outperform every other mobile app category in customer conversions. The median health and fitness app has a 35-day customer conversion rate of 2.7%, according to RevenueCat.
The top health and fitness subscription apps (those in the top 10% of performance), have a 35-day customer conversion rate of 12.1%—more than double the industry average.
Annual subscriptions dominate
In 2024, 41.4% of subscription apps primarily pushed annual subscriptions, compared to monthly subscriptions (29.8%) or weekly subscriptions (27%). There are some key benefits to annual subscriptions—developers get subscription money upfront, and they don’t have to ask customers to renew for 12 months.
Renewal periods coincide with an uptick in customer churn, as people make notes to themselves to cancel subscriptions that they’re not getting sufficient value from, according to RevenueCat.
The primary benefit of weekly and monthly subscriptions is a lower upfront cost, which can help convert customers who don’t want to pay $20 (and in some cases much more) for an annual subscription to an app.
Annual subscriptions are most popular with health and fitness apps. Roughly 2 out of 3 health and fitness apps primarily offer annual subscriptions, with a median price of $40 per year.
Monthly subscriptions are most popular with shopping apps. Some 51% of shopping apps primarily offer monthly subscriptions, with a median price of $5.99.
Weekly subscriptions are most popular with mobile games. Nearly 8 out of 10 games offer weekly subscriptions, with a median price of $4.99.
Across all annual subscription plans, education apps have the highest median annual subscription fee, at $44.99 per year. Travel apps have the lowest median annual subscription fee, at $18.67 per year, according to RevenueCat.
Dynamic offers based on user behavior
Gone are the days of one-size-fits-all subscription pricing—today, the price you see inside an app is increasingly dependent on your location and behavior.
For example, a highly engaged user who signs up for a one-week trial period and checks into the app every day might see an offer for a full-priced annual subscription on day 3.
In contrast, someone who downloads an app, fails to sign up for the trial and skips a few days before opening the app again might see an offer for a monthly subscription, with a discount on the first month.
In some cases, audience segmentation happens immediately after download, as users get a questionnaire asking them to self-identify what they want from the app. Their answers can put them on two or more customer journeys inside the app, with personalized subscription offers.
Developers also offer different prices based on a user’s geography. The median cost of an annual subscription in North America is $35.99, while in Southeast Asia it’s $17.96, according to RevenueCat.
The iOS App Store still drives the majority of subscription revenue, with 76.1% of developers saying they earned $8 out of $10 (or more) subscription dollars from Apple downloads.
Hybrid monetization
Successful apps combine multiple monetization streams, offering subscriptions, in-app purchases, and ads.
Mobile games are most likely to embrace hybrid monetization, with roughly 6 out of 10 subscription-based games also offering in-app purchases, lifetime subscriptions, and other monetization engines.
If your mobile app isn’t taking a hybrid approach to monetization, you could be leaving money on the table. It’s important to test a monetization mix that works for your customers and your specific app. Depending on your app, some of your customers might prefer a lightweight, ad-supported version, while others will be happy to buy a subscription for the full feature set.
Monetize your mobile apps with ads from Start.io. Connect with us to learn more about how Start.io can help.
Mobile games saw 49.3 billion downloads in 2024—a decline of 6.5 percent from 2023, and 14 percent lower than an all-time high set in 2020, according to new data from Sensor Tower.
Despite the decline in downloads, mobile games saw year-over-year improvements in both in-app revenue and time spent playing games, suggesting that mobile publishers are building stickier games, and getting better about extracting revenue from games.
In 2024, people spent an estimated 390 billion hours playing mobile games, roughly 7.7 percent higher than time spent gaming the year before. Mobile games pulled in an estimated $81.7 billion in in-app revenue last year, a little over $3 billion more than 2023, according to Sensor Tower.
More than 60 percent of the world’s population lives in Asia, so it’s no surprise that Asia leads the world in mobile game downloads (21 billion) and in-app revenue ($38 billion). But growth stalled last year in Asia, which saw a 3 percent year-over-year decline in revenue, and a 6 percent year-over-year decline in downloads.
On a country level, downloads and in-app revenue were either flat or down in Asia’s four biggest mobile markets: China, Japan, South Korea, and Taiwan.
The United States is the single biggest country on the planet for both mobile game downloads and in-app revenue, generating $25.8 billion in mobile game revenue alone in 2024—or roughly 1 out of every 3 mobile dollars worldwide.
In-app mobile game revenue grew in the U.S. by 8.8 percent year-over-year, while mobile game downloads declined by around 10 percent.
Scopely’s 2023 breakout hit Monopoly Go! was the most lucrative mobile game of the year, bringing in more than $2.5 billion alone.
The most successful new game launch of the year was Pokemon TCG Pocket, which generated an estimated $5.3 million per day—roughly equivalent to $1.9 billion per year if its revenue remains constant.
Chinese mobile giant Tencent is the largest mobile game publisher of the year, bringing in more than $10 billion in revenue last year, according to Sensor Tower. Its next-closest competitor, Scopely, pulled in an estimated $3 billion.
Tinder remains the most-downloaded dating app in 45 out of 50 U.S. states, driving more than half of its parent company’s in-app revenue in 2025, according to a new data analysis from Start.io.
Tinder’s next closest competitor, Bumble, lost relative market share over the past year, amid a CEO change and declining company’s stock value, according to Start.io’s analysis.
This year, Tinder and Bumble both saw fresh competition in the U.S. from the Czech company FlintCast, makers of the dating apps Evermatch, SweetMeet, Maybe You, and iHappy.
For this analysis, Start.io used its privacy-compliant, anonymized first-party audience segmentation data to study more than 400,000 active mobile devices in the United States that have one or more dating apps installed. This data includes mobile app installation rates, but does not include whether someone is actively using the app.
Another banner year for Tinder
Match Group owns more than 60 dating websites and apps, but one app stands above them all: Tinder. The dating app that popularized the phrase “swipe right” generated $1.94 billion in revenue for Match Group in 2024—a little more than half of the company’s total revenue.
And while Tinder has tens of millions of users, it makes money from the roughly 9.7 million people who each pay the company an average of $16.68 per month to use the app, according to Match Group’s FY 2024 annual report.
In 2025—13 years after the app initially launched—Tinder is the most-installed dating app in 45 U.S. states, according to data from Start.io. Tinder is #2 in Arkansas and Minnesota, and #3 in Louisiana, Rhode Island, and South Dakota.
Match Group’s other apps
Match Group, which employs around 2,500 people and is valued at around $8 billion, runs a number of smaller dating sites and apps—none of which come close to Tinder’s audience in size.
In Start.io’s analysis of the top 10 most-downloaded dating apps in each U.S. state, seven Match Group-owned apps made the cut:
- Tinder: Top 10 in 50 states
- Plenty of Fish: Top 10 in 40 states
- Hinge: Top 10 in 35 states
- OkCupid: Top 10 in 5 states
- BLK: Top 10 in 3 states
- Chispa: Top 10 in 3 states
- Match: Top 10 in 2 states
Collectively, Match Group’s non-Tinder properties generated $1.47 billion in in-app revenue in 2024.
Bumble stumbled in 2024
2024 was a challenging year for dating giant Bumble, owner of mobile apps Bumble, Bumble for Friends, Badoo, and Geneva. The company hired a new CEO in early 2024, who resigned in early 2025.
Bumble cut about 30 percent of its employees in 2024, and lost its chief financial officer, chief business officer, and chief technology officer in 2025, according to its latest annual report. The company is shutting down Fruitz, a Gen Z-focused dating app it acquired in 2022, and Official, an app it acquired in 2023.
Bumble’s stock reached an all-time low in 2025, falling from around $75 per share at its 2021 IPO to less than $5 per share in March 2025. Wall Street currently values the company around $530 million.
For years, Bumble was unique among dating apps because it didn’t allow men to send the opening message to women in the app’s heterosexual dating section. This solved a persistent problem in heterosexual dating apps, of men flooding women’s dating inboxes with generic, occasionally hostile, opening messages. In 2024, Bumble changed that core feature, inviting men to respond to an Opening Move from women they matched with.
Bumble ranks among the top 10 most-downloaded dating apps in every state in the U.S., and the #2 most-downloaded dating app in 24 out of 50 states (in most cases, behind #1 Tinder), according to Start.io’s analysis. The company’s other top dating app, Badoo, is in the top 10 dating apps in 47 out of 50 states.
Grindr remains steady
Grindr, which specializes in gay, bi, trans, and queer dating, stayed relatively constant between 2024 and 2025, appearing on the top 10 most-downloaded dating app list in 45 out of 50 U.S. states, according to Start.io data.
Grindr is the #3 most-downloaded dating app in two states—Arkansas and New Mexico, according to Start.io data.
Grindr ended 2024 with an average of 14.2 million monthly active users, a nearly 7 percent increase in MAUs from 2023, according to the company’s latest annual report. Grindr’s stock price more than doubled in 2024, starting the year at around $8.30 per share, and ending the year at roughly $18 per share. Wall Street currently values Grindr at around $3 billion.
New upstarts trying to squeeze into the dating space
Czech Republic-based FlintCast had a big year for growth in the U.S., gaining a foothold with their apps SweetMeet, iHappy, Maybe You, and Evermatch.
These four apps failed to crack the list of top 10 most-downloaded dating apps in Start.io’s 2024 analysis. Not so today:
- SweetMeet: Top 10 in 50 states
- iHappy: Top 10 in 46 states
- Maybe You: Top 10 in 37 states
- Evermatch: Top 10 in 13 states
Other notable privately owned dating apps in 2025 include:
- Boo: Top 10 in 30 states
- Happn: Top 10 in 10 states
- Hily: Top 10 in 10 states
How often do you see ads inside your favorite mobile app? Today, there’s a good chance the answer depends on how you interact with the app.
Mobile game developers are experts in audience segmentation, building multiple experience paths for their users based on their characteristics. Mobile audience segmentation can lead to revenue uplift of 20 percent or more in average customer lifetime value, according to the panelists at a recent monetization talk at Gamesforum Barcelona.
Defining audience segmentation
Audience segmentation is the process of organizing a group of people into groups that share common characteristics.
Brands use audience segmentation to send ads to the people most likely to purchase their goods and services. Advertising platforms like Start.io offer advertisers thousands of audience segments, from the very large (“everyone in the United States,” “women globally between the ages of 18-34”) to the very small (“commercial goat farmers in Idaho,” “bicyclists in Hawaii”).
Mobile app developers look at audience segmentation through a different lens: Their audience is everyone who uses their app, and audience segmentation is the process through which they group their audience into two or more buckets, based on common characteristics.
For mobile app developers, the goal of audience segmentation is to increase customer lifetime value by doing two things:
- Extending the time people use the app before abandoning it
- Showing people the monetization levers they’re most likely to interact with
For large developers, small improvements to monetization can mean millions of dollars in additional revenue over time.
“We are very happy with a small uplift,” one game developer said at Gamesforum Barcelona. “Sometimes we have 5 percent uplift, sometimes we have 3 percent uplift. Sometimes we just get insights. It’s quite doable to get 15- to 20 percent overall [revenue per user] uplift—it’s quite doable, just start doing it.”
Audience segmentation through spending behavior
For more than a decade, mobile game developers have used a fishing analogy to segment their users by spending behavior—big spenders are “whales,” medium spenders are “dolphins,” and small spenders are “minnows.”
“Minnows” might see lots of rewarded videos, where they can earn free in-game items for watching ads. On the other side, “whales” might see very few ads, and instead encounter frequent offers to buy high-ticket items with cash.
Both groups are happier with a personalized experience: Minnows don’t mind spending their time, and whales don’t mind spending their money.
With a behavior-based segmentation path, all users get the same experience at the start of the game, with tests in early gameplay to learn whether they’re whales, dolphins, or minnows.
Audience segmentation through user acquisition channel
For some mobile app developers, audience segmentation begins with one of the earliest signals that a user generates—whether they downloaded an app organically, or through an ad campaign.
One mobile game company that earns $200 million per year from one of their main titles found that people who download the game through a paid media campaign are typically skeptical of the game at first, and need a few days of uninterrupted gameplay to get hooked, a developer at Gamesforum Barcelona said.
If the person is still playing by day 3, the company gradually turns on ads and offers for in-game items. This developer found that people who downloaded their game organically had far less skepticism, and were OK with seeing ads and in-game offers on day 1.
Audience segmentation through in-app decisions
Mobile app developers have a tough challenge—they need to design apps that are quick and easy to consume for casual users, and richly engaging for hardcore users.
Mobile game developers segment their audience into casual vs. hardcore players, by measuring how often someone opens the game, how long they play during an active session, and what types of decisions they make inside the game.
At Gamesforum Barcelona, one game developer described creating up to six audience segments based on user behavior inside the game.
On one end of the spectrum, ultra-casual players typically open the game sporadically, play for a few minutes and don’t open the game again for multiple days. This group typically doesn’t like engaging with rewarded video ads, and will close the app early if they encounter an ad they can’t skip.
On the other end of the spectrum, ultra-hardcore players open the game several times a day, and play longer game sessions. This group is most open to rewarded video ads, and more likely to buy in-game items.
Audience segmentation through demographic characteristics
Game developers collect (or infer) demographic information about their players, and use this information to segment their audience and personalize the in-game experience.
This could include characteristics like age, gender, geographic location, and interests.
How to set up audience segmentation tests
At Gamesforum Barcelona, game developers said they are constantly building monetization and audience segmentation tests that run on 10- to 20 percent of the user base. If the test is successful, it’s gradually introduced to the entire user base.
Tests are designed to answer two questions:
- Is this a better way to segment our audience?
- Does this test yield higher average lifetime value per user?
There are a nearly infinite number of ways to segment an audience, and A/B tests reveal which segments yield meaningful differences in user behavior.
Could your mobile app benefit from audience segmentation?
Generative AI apps broke the 1 billion milestone in 2024, with 1.49 billion cumulative downloads worldwide, and nearly $1.3 billion in revenue from in-app purchases, according to new data from Sensor Tower.
To put that growth in perspective, generative AI apps were virtually nonexistent just two years ago. In 2022, generative AI apps pulled in just 119 million downloads worldwide and $30.1 million from in-app revenue. The market is now 12x bigger for downloads, and 42x bigger for in-app revenue.
Mobile developers are integrating AI into their apps (or saying they are) in huge numbers. Apps that mentioned “AI” in their name, subtitle, or app description were downloaded 17 billion times in 2024, according to Sensor Tower.
ChatGPT is far and away the most popular generative AI app on the market, leading in both downloads and in-app purchases in every major country globally in 2024 except China, where it’s not available. When ChatGPT launched its mobile app in May 2023, it took just five months for the company to reach 50 million monthly active users (MAU), which represents one of the fastest mobile adoption rates in history.
By comparison, Temu, Disney+ and Reddit all took longer to reach 50 million MAU—5 months for Temu, 12 months for Disney+ and 62 months for Reddit.
People spent an estimated 7.7 billion hours chatting with AI bots in 2024, an increase of 347 percent from the previous year, according to Sensor Tower.
India leads mobile downloads globally
Zooming out to a global level, India remains the undisputed world leader in mobile app downloads, while Americans spend more money on in-app purchases than any other country,
Indians downloaded an estimated 24.36 billion mobile apps and spent 1.1 trillion hours on their phones in 2024, according to Sensor Tower.
With an estimated 650 million smartphone users in the country, that means the average Indian mobile user downloaded 37 apps and spent 1,700 hours on their phones last year. More than half of smartphone users in India are under the age of 24, according to Start.io data.
The United States is just as mobile-obsessed as India, with Americans downloading an estimated 12.3 billion mobile apps and spending 323 billion hours on their phones, according to Sensor Tower.
Back-of-the-napkin math: With an estimated 300 million smartphone users in the U.S. in 2024, that means the average American downloaded 41 mobile apps, and spent 1,075 hours on their phones.
Where the United States really shines is in mobile in-app purchases, with Americans spending an estimated $52.4 billion inside apps in 2024, according to Sensor Tower. That’s around $175 per person, per year.
Globally, people spent an estimated $150 billion on mobile in-app purchases in 2024, with about a third of that revenue coming from the U.S. alone.
Mobile games generate more in-app revenue than any other category by far. In 2024, people spent about $80 billion in-app on mobile games, and $69.2 billion on all other types of apps.
Video apps—a category that includes apps like like Disney+, Max, and WeTV—pulled in $12 billion in mobile subscription revenue in 2024, followed closely behind by in-app spending inside social media apps.
In 2024, five mobile apps crossed into the mobile monetization hall of fame, generating more than $1 billion in mobile in-app revenue in a single year. These included the games Brawl Stars, Dungeon & Fighter, Last War, and Whiteout Survival, and the Chinese streaming video app WeTV.
There are now 26 mobile apps that have managed to generate more than $1 billion in mobile, in-app revenue in a single year, according to Sensor Tower:
- Clash of Clans (2014 or earlier)
- Puzzle & Dragons (2014 or earlier)
- Monster Strike (2015)
- Honor of Kings (2017)
- Netflix (2018)
- Fate/Grand Order (2018)
- Tinder (2019)
- Coin Master (2020)
- Game for Peace (2020)
- Pokemon Go (2020)
- PUBG Mobile (2020)
- Roblox (2020)
- TikTok (2020)
- Candy Crush Saga (2021)
- Disney+ (2021)
- Garena Free Fire (2021)
- Genshin Impact (2021)
- YouTube (2021)
- Google One (2022)
- Monopoly Go (2023)
- Royal Match (2023)
- Brawl Stars (2024)
- Dungeon & Fighter (2024)
- Last War (2024)
- WeTV (2024)
- Whiteout Survival (2024)
In the fast-evolving world of programmatic advertising, efficiency and transparency are key to maximizing revenue for publishers. One of the most widely adopted solutions for header bidding is Prebid, an open-source framework that enables publishers to integrate multiple demand sources and optimize competition for their ad inventory.
At Start.io, we’re committed to ensuring seamless participation in the programmatic ecosystem by developing Prebid adapters—both client-side (Prebid.js) and server-side (Prebid Server). In this article, we’ll explore what Prebid is, why it’s crucial for modern ad monetization, and the differences between client-side and server-side integrations.
What is Prebid?
Prebid is an open-source set of software solutions that simplify header bidding, allowing publishers to connect with multiple demand partners in a fair, real-time auction before making an ad request to their primary ad server.
By leveraging Prebid, publishers increase competition for their ad inventory, achieve higher eCPMs, reduce latency and improve page load speeds, and gain access to analytics and auction insights.
Prebid offers two main ways to integrate with demand sources: client-side (Prebid.js) or server-side (Prebid Server). Each offers its own advantages, depending on the publisher’s needs.
Client-side vs. server-side Prebid adapters: Key differences
As we develop adapters for both Prebid.js (for client-side auctions) and Prebid Server (for server-side auctions), it’s important to understand how they differ.
| Feature | Prebid.js | Prebid Server |
| Auction location | Runs in the user’s browser | Runs on a separate server |
| Latency | Can be slower due to multiple network calls from the browser | Faster, as bidding happens server-side |
| Transparency | Publishers have full visibility into bids | Less transparent due to server-side processing |
| User syncing | More control over cookies and user data | Limited access to cookies due to privacy restrictions |
| Implementation complexity | Easier to set up | Requires infrastructure for server management |
| Data privacy | Subject to browser limitations (ITP, Privacy Sandbox, etc.) | Less affected by browser restrictions |
What is Prebid.js?
Prebid.js runs directly in the user’s browser, sending bid requests from the publisher’s page to multiple demand sources. The browser selects the highest bid and passes it to the ad server for final decisioning.
- Pros: Higher transparency, better cookie matching, easy set up
- Cons: Increased latency, browser restrictions (e.g. Apple ITP, Google Privacy Sandbox)
Server-side Prebid
In contrast, Prebid Server offloads the auction process to a server, reducing the number of network calls from the browser. Instead of handling bids in the client environment, the request is sent to Prebid Server, which communicates with multiple demand partners and returns the highest bid.
- Pros: Lower latency, better scalability, reduced impact from browser privacy restrictions
- Cons: Reduced transparency, weaker user ID matching, requires server infrastructure
Why Start.io is building Prebid Adapters
At Start.io, we are laser-focused on helping publishers maximize their ad revenue opportunities, across both client-side and server-side environments.
By developing Prebid.js adapters, we’re helping publishers participate in real-time auctions directly within the browser. Meanwhile, our Prebid Server adapter helps publishers who would rather participate in real-time auctions using server infrastructure.
By supporting both integration methods, we’re providing flexibility for publishers to choose the right setup based on their needs: Prioritizing transparency with client-side bidding, or speed and efficiency with server-side auctions.
Prebid has become a cornerstone of programmatic ad monetization, offering publishers greater control, competition, and transparency in header bidding. Understanding the differences between Prebid.js and Prebid Server is crucial for choosing the right approach to ad monetization.
Stay tuned for more updates as Start.io expands its capabilities in the Prebid ecosystem. Contact us today to learn more about how we can help you optimize your ad monetization strategy with Prebid adapters.
Start.io’s primary mission is to help mobile app developers earn the most money possible from the ads they run inside their apps. This month, we made a breakthrough in this effort, achieving 9.8 percent higher advertising fill rates and 12 percent higher daily revenue through our new partnership with Intent IQ.
Here’s how.
Every time Start.io needs to display an ad, it attaches (where allowed) the consumer’s mobile advertising ID, or MAID. MAIDs are a 32-character, anonymous and resettable alphanumeric identifier that gets attached to every mobile device.
For example, here’s what a sample MAID looks like: 34868ee1-3299-465e-a776-4fe0abd869c6
Adtech companies attach consumer demographic information to MAIDs, like the consumer’s general location, their interests, estimated income, and other factors, based on what companies can observe (or guess) about the consumer’s profile.
When we send an ad request that includes a MAID, advertisers compete with one another to show an ad to that person, based on the demographic information attached to their MAID.
For example, car companies pay high CPMs to show ads to people actively shopping for their next car. On the flip side, a large potato chip company might pay low CPMs to reach everyone in America as part of a general brand awareness campaign.
MAIDs have some limitations. For one, consumers can reset their MAID, or choose not to share their MAID with advertisers. Google and Apple can (and have) restricted the industry’s ability to use MAIDs for ad targeting. Regulators worldwide have signaled that they could enact new rules around how companies use industry-standard advertising identifiers like MAIDs and their browser equivalent, third-party cookies.
Enter alternative advertising IDs.
These IDs work like MAIDs and serve as a repository for companies to organize consumer demographic and behavioral information. Intent IQ’s identity resolution platform helps companies like Start.io match MAIDs with one or more corresponding alternative IDs. Intent IQ maintains its own alternative ID, IIQ ID, and helps companies match MAIDs with other third-party alternative IDs.
With Intent IQ, we’re now able to send more information along with our advertising requests—ideally, both a MAID and one or more of the MAID’s corresponding alternative IDs.
In our tests, we used Intent IQ’s platform to match MAIDs with their corresponding alternative IDs. We then sent out millions of real-world ad requests and compared the performance of enhanced ad requests (containing multiple advertising identifiers), with a control group (containing only MAIDs).
In an industry as hyper-optimized as ours, winning 12 percent higher daily revenue is a major win. Start.io delivers hundreds of millions of ads per day to more than 500,000 active mobile apps, so the impact of this test is both immediate and global.
We’re now in the process of introducing enhanced ads to all our demand partners.
Alternative IDs help the programmatic advertising ecosystem deliver ads that are better tailored to a consumer’s individual interests.
Everyone wins: Consumers get privacy-friendly ads that are better targeted to their interests, brands get their message in front of potential customers, and publishers get higher CPMs for their advertising inventory.
