10,000 startups later: what we learned about pivots

By PivotRadar Research • June 2025

---

Table of Contents

1. Executive Summary
2. Pivots: What is it all about?
3. First look at the data
4. Why data on pivots is hard to come by and analyze
5. Blockbuster pivots
6. Outlook

Executive Summary (AI)

Pivoting plays a crucial but often misunderstood role in startup success—VCs like Floodgate credit most of their returns to companies that changed direction. In our initial dataset of 10,500 “indie-type” startups, over 20% show signs of pivoting, while 21% have shut down—some with domains now repurposed for unrelated content like gambling sites. As Lean Startup principles spread and tech enables faster adaptation, pivoting appears to become more common. Yet tracking it remains challenging.

Pivots: What is it all about?

Pivots are “structured course corrections designed to test a new fundamental hypothesis about the product, business model, and engine of growth” (Ries, 2011). Pivoting decisions are “poorly understood”, says a research paper on entrepreneurial pivoting. Another study finds that liquidity and other constraints affect how pivots can play out. Also, pivots are not easy as, for instance, technical debt (such as existing software that is no longer ideal) or organizational limitations (such as lack of skilled staff to cater to a new market) need to be overcome. Yet, the magic of pivots is exemplified by a statement of Mike Maples of Floodgate, a VC, who wrote that “85% of Floodgate’s exit profits came from “pivots” – namely startups that started with one idea and then ended up course-correcting to something very different.” Similarly, Jeff Morris Jr., Managing Partner at Chapter One and former VP at Tinder, writes: “We romanticize visionaries who never waver, who push through resistance with stubborn certainty. But that narrative is increasingly obsolete.”

We’re kicking off this blog with a quick (and sometimes quirky) behind-the-scenes look at the early days of our initial pivot dataset—a collection of around 10,500 startups since 2014 that were featured (on a weekly basis, see the example here) at some point on platforms like Product Hunt, Indie Hackers, or in seed funding announcements (for example, from Techstars starting in 2017).

(Please note: this particular dataset is heavily skewed toward Product Hunt, Indie Hackers, Hacker News, Microlaunch, and similar platforms. In other words, it covers that grey area between indie projects and full-fledged startups. For deeper analysis on more established startups, you’ll occasionally find insights in our PivotRadar newsletter—and more extensively in the premium plans available at PivotRadar)

First look at the data

Our initial analysis shows that over 20% of these startups went through a clear pivot—meaning they changed their product, business model, or target audience. Meanwhile, 21% of them are no longer in business, and 5% of those defunct startups had their domains repurposed, often into spammy sites like gambling blogs.

Caveat: Startups were only classified as “acquired” if we found direct confirmation of the acquisition on their website. As a result, this number likely underestimates the true figure

The following chart shows the proportion of pivots among all tracked startups founded in the respective year and that are still active. The overall trendline is clear: pivoting has become more common. This may be due to the growing popularity of Lean Startup principles or the increasing ease of changing product direction in tech.

Why data on pivots is hard to come by and analyze

Aside from the fact that Eric Ries alone identifies at least 10 different types of pivots, analyzing the actual state of a startup is far from straightforward.

Take homer.co, for example. Back in 2020, their pitch was: “More easily discover apps via your friends.” Today, the homepage says: “Homer – Home Management App for homeowners: access any documentation about your home, anywhere and anytime.” Seems like a clear pivot, right?

But dig a little deeper and it gets messy: the signup/login doesn’t work (not even with Gmail addresses), and both App Store links are dead. So while Homer technically pivoted at some point, it’s basically dead now.

These kinds of ambiguous cases are common. Sometimes it’s even hard to tell if a site is now a niche blog—or just a placeholder landing page.

Then there are the even odder ones, such as arcopypaste.app. At first glance, the animated demo looks legit. But then you read the site’s text and find:

“Taking a screenshot of a Casino neu logo for example and blend it in with surrounding digital objects. This makes it much easier for casino brand creation.”

So… yeah. Another repurposed domain, now basically a gambling site. What it used to be? Clipdrop – Copy and paste reality with AR + ML.

Then there are the complete about-faces. Take the collage below as an example: this startup originally positioned itself around “learn as you invest,” but now presents itself as your personal “money manager”—promising a hands-off experience where you no longer need to do anything yourself (note: you better check the site’s Reddit reviews before “investing”).

In other cases, changes are initially subtle — especially to an outside observer. That’s where AI-assisted analysis can be particularly useful. Take, for example, the case of Glimpse, which rebranded to Panoplai. The image below shows the original website glimpsehere.com at the top, and the rebranded panoplai.com below. It becomes evident that Glimpse/Panoplai’s current offering has expanded beyond the original focus (i.e. a research platform that focused on language, emotion, and sentiment analysis) — suggesting a broader range of services and a potential strategic shift.

Different platforms

Remember when we wrote above that the data set skews heavily towards the more “indie” platforms such as Producthunt or Indiehackers? To see why this is relevant, we can compare the percentage of startups that have pivoted for two relatively different platforms: Techstars, an early-stage venture fund – and Producthunt, a community to discover startups, many of which relatively indie and one-person in nature. As the chart below shows, Techstars’ startups appear to pivot more, proportionally. Why? Difficult to say. Product Hunt founders have been said to often be more skilled in sales and marketing than in tech, so reconfiguring apps or servers after a pivot might be harder for them. Also, and in contrast to Techstars startups — typically formal companies whose name is expensive to change —Product Hunt entrepreneurs may simply launch a new domain for each new idea.

We can also see a small difference in the share of startup websites that went the “casino way”, i.e. were given up and have been repurposed by god-knows-who. 5.3% of ProductHunt featured startups have seen their website repurposed versus 4.5% for Techstars. What does such a repurposed domain look like? Well:

Blockbuster pivots

Of course there are absolute blockbuster pivots too. Many iconic companies began with entirely different ambitions before discovering unexpected success in a pivot. Hotmail started as a side project for a web-hosted database, but when VCs learned the founders had built their own email system to work privately, they were more interested in that—and Hotmail was born. Similarly, Slack and Flickr both originated from failed MMO game projects led by Stewart Butterfield, who recognized value in internal tools built during development. Other major pivots include Justin.tv/Twitch or Discord, which emerged from a failed multiplayer game, and Plaid, which shifted from a budgeting app to a $10B+ banking API platform after grappling with poor financial data access.

Outlook

This has been a short glimpse into what we will cover here from now on. Pivots are not just fascinating—they’re a data goldmine for investors and founders alike. That’s why, here at PivotRadar, we’re continuing to build out our tools and methods to analyze this often overlooked aspect of entrepreneurship. Check our queue for upcoming database entries here. Hope you enjoy.