Technical Due Diligence For Startups: Should VCs Care?
by Vlad Pavlov and Robert Garza
Avoidable Write Offs
In 2011 a Stanford computer science student Lucas Duplan founded a startup named Clinkle. It was supposed to develop a mobile payment app, similar to Apple or Google Pay.
Clinkle had a secret sauce: they wanted to use ultrasound waves for communication between a phone and a Point Of Sale device in a store (POS). Apple Pay and Google Pay took a different approach. These apps use Near Field Communication (NFC), which relies on short-range electromagnetic transitions, for connecting phones with POS devices.
Clinkle raised $30 M from a number of prominent investors. 4 years later the company was shut down. Another failed startup.
Most startups fail. It’s the nature of the startup business. However in case of Clinkle the basic assessment of technology feasibility could have prevented investors from pouring money into the company. After spending a day or two doing Technical Due Diligence (TDD), a reasonable engineer-generalist would discover that:
- Most smartphone models were not purpose-designed to use ultrasound;
- Even though some models had a capacity for ultrasonic communication, it was unreliable and insecure.
In other words, Clinkle’s technological roadmap would not pass TDD. But it seems that TDD was not a part of the process.
Clinkle appears to be an example of a legitimate startup, though with an overoptimistic founder. Pure fraud is another type of a startup failure where TDD can prevent investors from losing money.
In 2013 Shaukat Shamim founded YouPlus, which claimed that it had built “an advanced AI and Machine Learning platform to unlock consumer opinions and experience insights from videos”. Since its inception the company raised over $17 M. Later it was discovered that the company never developed an actual AI; instead it was paying workers in India to watch and manually label videos.
It would take a few days for an independent engineer with access to company employees and source code to discover that they did not have an actual technology. But it seems investors skipped TDD, so our hypothetical engineering expert never looked at the source code and was never able to advise VCs to pass on this investment opportunity.
Does Technology Matter?
How frequently and thoroughly do VCs practice TDD?
The authors were not able to find a comprehensive industry report or academic research focused on this subject. On a personal level, we have good connections with (and some understanding of the inner workings of) a couple dozen funds. To our knowledge, only two of them systematically perform TDD as a part of their overall Due Diligence (DD) process. Most of the funds focus on three major areas during DD:
- Founding team and executives
- Market and business model
- Legal status and financials
Technology-related questions often are covered by a one-or-two hour call with a startup’s CTO, without exhaustive technical examination.
On the surface, this distribution of time makes sense. Startups die frequently, but technology is rarely considered to be the cause of failure. In 2019 CBInsight analyzed over 100 startup failure postmortems. They identified the top 20 reasons startups die, which you can see on the chart below.
Technology-related problems did not make it to the Top-20 list. So, should VCs consider TDD an insignificant activity and avoid spending their time/money on it? No, it’s not that simple.
Let’s look again at Clinkle. When the company finally closed, the CEO wrote “we unfortunately could not find a strong enough consumer pain point so decided to discontinue the service”. Do you really believe that lack of “a strong enough consumer pain point” was an actual reason for the company’s failure? Of course not, otherwise Google Pay and Apple Pay would have never become as popular as they are now. Still, from the CEO’s perspective, the reason for Clinkle’s death was the first item in the chart above.
Another example. Until the very end of her tenure at Theranos, Elizabeth Holmes believed her R&D team would soon make their blood-testing technology work, she just needed the right people to execute her vision. She frequently replaced engineers who could not make it happen. When asked, she would blame reason # 3 “Not The Right Team” for her failure. But the industry experts are saying it was “physically impossible” to deliver on Theranos’ promise. We do not know whether Holmes was a visionary misled by her delusions, or a cold-blooded criminal. Maybe both. But we do know that conducting TDD could have saved Theranos investors $ 1.4 B.
In the cases of Clinkle, YouPlus, and Theranos, technology did matter. But their founders focused on other problems while talking to investors, and the investors bought into it. These examples capture a reason for why technology-related failures are not in the CBInsight’s Top-20 list. The creators of the list relied on self-assessment reports by the founders of closed startups, who could hardly be objective. If we conduct an independent in-depth post mortem and root cause analysis, the authors are convinced technology-related problems would make it to the Top-10.
Conclusion
Roses Are Red, The Sky Is Blue And TDD Prevents Write Offs
Technological issues were not the only (and perhaps not even the largest) warning signs at Clinkle, YouPlus or Theranos. Still, it seems like employing TDD could have been a very inexpensive way for their investors to uncover the existence of red flags. Which in turn, could have prevented them from losing money.
From the authors’ perspective, prudent advice for VCs is simple: always conduct TDD. It is a tiny additional investment that can prevent some very unfortunate write offs.
P.S. The authors would like to express our sincere gratitude to Alex Chachava, Max Schkolnik, Denis Dovgopoliy, Nick Mityushin and Dmytro Bidnyak, whose feedback from whom helped us shape the ideas presented in this post.
