More broadly there is this idea in tech startup internet bubble land that you just need a novel idea to get you to product market fit. In reality you need to execute. Starting a car company isn’t a bad idea simply because many competitors exist. But you better build great cars. If you don’t, you might be inclined to say “we couldn’t find product market fit”
I think this is close but not quite right. There are many "great" products that nobody wanted to buy. Sure building a great product gives you a chance to succeed, but building a product people want to buy is the real goal.
"Great" is also just too vague to be useful. Is a great car one that's cheaper than the rest? Better mileage? More dependable? More beautiful? More luxurious? Safer for kids to ride in? More fun to drive?
I've always liked the Donald Miller Storybrand framework for figuring out these kinds of puzzles:
https://www.amazon.com/Building-StoryBrand-Clarify-Message-C...
But they have product market fit, in that they operate in a huge market with plenty of suckers. They were good at sales. They could only keep a handful of customers at any one time. Plenty of sales stuff were straight lies - they'd make up clients, used an intern for testimonials, etc.
They're bootstrapped and still around. A VC commented that they were resilient, cockroach like, all the sexy words you use during a recession.
But it's likely they just never grew into a unicorn level because they kept bleeding customers.
https://www.npr.org/2022/12/27/1145616523/southwest-airlines...
And from the WSJ: https://archive.is/3ujIL
What about a lack of tests, or bad coding standards?
We don't know the particulars, but one can reasonably infer that the root causes were broader and more systemic than that (i.e. a mix of technical and organizational issues).
https://www.sec.gov/litigation/admin/2013/34-70694.pdf
> On August 1, 2012, Knight Capital Americas LLC (“Knight”) experienced a significant error in the operation of its automated routing system for equity orders, known as SMARS. [...]
> Upon deployment, the new RLP code in SMARS was intended to replace unused code in the relevant portion of the order router. This unused code previously had been used for functionality called “Power Peg,” which Knight had discontinued using many years earlier. Despite the lack of use, the Power Peg functionality remained present and callable at the time of the RLP deployment. The new RLP code also repurposed a flag that was formerly used to activate the Power Peg code. Knight intended to delete the Power Peg code so that when this flag was set to “yes,” the new RLP functionality—rather than Power Peg—would be engaged."
> [...] In 2003, Knight ceased using the Power Peg functionality. In 2005, Knight moved the tracking of cumulative shares function in the Power Peg code to an earlier point in the SMARS code sequence. Knight did not retest the Power Peg code after moving the cumulative quantity function to determine whether Power Peg would still function correctly if called.
A system that can, in a catastrophic failure, single-handedly put your entire company out of business, is what I'd consider mission-critical.
Mission-critical systems should, by definition, be held to a higher standard of quality than non-critical systems.
A mission-critical system that has dead code still in production for 9 years, untested, and allowing the "repurpos[ing of] a flag," suggests the level of quality that may be at play.
Your guess is as good as mine, but this smells like a pretty grievous coding error, lack of quality around deployment processes, and insufficient testing and validation.
For a system that is so mission-critical, that a simple "mistake" put the entire company out of business in one single event, this seems pretty "horrible" to me.
YMMV.
It didn't actually kill the company, just the product line.
Fun fact: Ashton was actually George Tate's parrot, living in a large cage in the L.A. office.
> Ma.gnolia servers lost all data in a complete outage on January 30, 2009.On February 17, Halff announced that due to data corruption, all user data in the database was irretrievable, rendering the site essentially dead.
