Strava Stays Strong Against Scrapers Ahead of IPO

AI companies have transformed into data-centric entities, creating a heightened demand for increasingly larger datasets for training purposes. In response to this demand, many AI startups are sidestepping crucial internet protocols—such as adhering to robots.txt files that dictate which sections of a site should remain off-limits to automated crawlers—opting for more aggressive data scraping tactics. Consequently, numerous websites have begun to restrict data access and, in some cases, are entering discussions about licensing agreements with AI firms. Strava, a platform dedicated to fitness and social running, is one such entity that is adopting this strategy by imposing restrictions on access to its site and initiating fees for developer utilization.

To address scraping concerns, Strava is bolstering its platform’s security and will limit the accessibility of specific data to authenticated users only. Previously, public profiles and fitness club listings were viewable without logging in. Now, users must authenticate to access this information, effectively minimizing unauthorized scraping by AI.

As for the API, developers once enjoyed a free, tiered access model for Strava’s platform—starting with basic access and later increasing it as their application expanded. However, the company is now instituting a standard fee of $11.99 per month for all developers, with the possibility of regional price variations.

Strava has reported an increase in its developer community, growing from 185,000 last year to 241,000 this year, and the company remains committed to supporting them. As part of this initiative, Strava plans to endorse the Model Context Protocol (MCP), a new standard that will enable AI assistants and applications to systematically access external data, granting Strava enhanced control over information sharing.

Additionally, the company intends to disable specific API endpoints—particular channels through which external applications access designated data, such as club information—to protect user data. In 2024, Strava tightened its API regulations, limiting its utility for AI training and imposing restrictions on third-party applications that display data from other users. These changes were met with pushback from developers who expressed concerns about the potential adverse effects on their applications.

While some developers might be amenable to the subscription model, the removal of certain API endpoints could still negatively impact applications that depend on them. Strava is providing developers with a 90-day grace period before these changes take effect.

In an interview with TechCrunch, Strava’s CEO, Michael Martin, shared concerns that rampant AI scraping could endanger the future of the public internet.

“AI companies are aggressively scraping public websites to satisfy their insatiable need for training data, which universally degrades site performance,” Martin remarked. “We have recently encountered numerous instances where performance has significantly decreased. In addition to scraping public sites, they are also trying to access our data through our API, ignoring our API terms.”

He disclosed that Strava has declined requests for data licensing agreements from major AI laboratories. Martin specifically referenced Perplexity, noting that the AI search startup attempted to shield its scraping activities through aggregator services to mask its source, despite receiving a denial. This mirrors Perplexity’s previous actions in similar situations.

Martin also pointed out concerns related to server overloads caused by inefficient API requests from collaborative applications, which often place undue stress on Strava’s infrastructure. This issue resembles Meta’s reasoning for banning third-party chatbots from WhatsApp last year, citing system overload challenges.

The timing of these measures appears deliberate. Strava confidentially filed for an IPO earlier this year, and its data protection strategies may be designed to demonstrate data stewardship to potential investors. Martin quickly acknowledged comparisons to Reddit’s API access limitations from 2024, clarifying that, unlike Reddit—which based API access fees on call volume, making it prohibitively expensive for many developers—Strava’s flat fee structure aims to foster a thriving developer ecosystem.

“We want our users to feel they own their data and be confident in how we handle and protect it. At the same time, we are dedicated to ensuring that developers continue to thrive and expand,” Martin concluded.