Tech VCs, to their own detriment, are increasingly investing in businesses that aren't software-driven
One sign that technology markets often exhibit at the tail end of a cycle is a fast diversification of the types of startups getting funded. For example, following the core internet boom … Tweets: @eladgil , @mdudas , @ganeumann , @asanwal , @shaig , @zachcoelius , @kimmaicutler and @semil Tweets: Elad Gil / @eladgil : End of Cycle? “A software-enabled, network connected, smart toaster is, when all is said and done, just a toaster” http://blog.eladgil.com/... Mike Dudas / @mdudas : “We are at the end of an economic cycle for tech, and tech investors are desperate for the next new thing.” @eladgil http://blog.eladgil.com/... Jerry Neumann / @ganeumann : Completely agree: in despration, “tech investors are investing in...industries they know nothing about” http://blog.eladgil.com/... Anand Sanwal / @asanwal : A software-enabled, network connected, crowd funded, smart toaster is still just a toaster. http://blog.eladgil.com/... gold by @eladgil @shaig : “tech investors are now investing in areas they do not understand, at valuation multiples that do not make sense” http://blog.eladgil.com/... Zach Coelius / @zachcoelius : Smart thinking from @eladgil about the current state of the tech investment cycle https://lnkd.in/ePrfmSH Kim-Mai Cutler / @kimmaicutler : Investors are applying tech multiples to radically different types of (non-software) cos. It's unlikely to end well. https://twitter.com/... @semil : Excellent by @eladgil on potential perils of excess $ searching for the next thing, with a historical perspective: http://blog.eladgil.com/...
Context & Ripple Effects
A year after Andreessen Horowitz made its public case against bubble talk, Elad Gil's post supplies the bear-side ledger: when a funding wave stops compounding inside software and starts spilling into hardware and consumer products, the diversification itself is the late-cycle tell. His line that a connected smart toaster is still just a toaster captures the mispricing mechanism — tech multiples applied to businesses without software economics.
The subsequent record bears him out. Core categories were squeezed first, with early-stage rounds worldwide falling from roughly 13.3K in 2014 to about 5.9K by 2017, hitting mobile app and SaaS hardest, while capital chased whatever was next.
First-order effects
- Funds deploying into non-software companies at software-style valuations are carrying mispricing risk on precisely the assets least able to defend those multiples once growth normalizes.
- Founders in core mobile and SaaS categories now compete for scarce partner attention against trendier hardware and consumer bets — the squeeze that shows up in the 2017 round-count collapse.
Second-order effects
- When the cycle turns, non-software holdings get marked down ahead of software ones, pushing general partners back toward familiar categories or the next theme — the overpaying-then-pivot dynamic later visible in the 2021 survey of VC overpayment and Web3 bets.
Third-order effects
- If the pattern holds across cycles, each boom ends the same way this one did: a macro-forced reckoning with inflated valuations and overzealous funds that resets which categories get funded at all, while the underlying shift of software fading into background infrastructure keeps lowering the bar for what counts as a 'tech' company.
The trend: Late-cycle venture capital reliably migrates from its core software franchise into adjacent physical and consumer businesses, making that diversification one of the more dependable top-of-market signals.