Why tech responds poorly to crises, and how to do better
Sun, Jun 9, 2013 at 2:00 PM by Dave Winer.
  • A picture named tylenol.gifThe first time I saw a tech company blow it in the Internet age was in 1994 when Intel was trying to quell public outrage about a problem with math functions on their Pentium chip. According to Wikipedia, an estimated 1 in 9 billion floating point divides would produce inaccurate results. The problem could be demonstrated in Excel. Intel said the flaw was so small that it didn't warrant any concern by users. They were probably right. But that didn't stop the outrage from escalating. Every time Intel spoke, the problem got worse. Eventually they had to offer replacement chips to any user who wanted one. It might have cost much less if they had admitted the problem at the start.
  • It doesn't happen often that the press sides with users, but when it does, the tech industry usually reacts poorly. The reason is simple. They're accustomed to Gee Whiz treatment from the press. That the people who run the companies are themselves miracles. Because the products they make are so impressive to the people in the press.
  • But that's going to fade over time, as tech products become more ordinary. Kids who were brought up with the products don't think they're so amazing, and they'll become reporters or bloggers, and the tech industry will have to deal with crisis not by stonewalling, but with empathy, and understanding of how the public thinks.
  • An exec at Google or Facebook might be puzzled by the reaction to the NSA news. Didn't they already know that we have to provide the government with information when they have a legal basis for requiring it? Maybe people did know. But that doesn't matter. Right now they're not happy about it. And the blame is falling on the tech companies. Usually reporters just rewrite press releases. But every once in a while they express their independence. This is one of those times.
  • The classic example of a company responding well to a crisis that's no fault of theirs was Johnson & Johnson with the Tylenol murders. If they had spun it the way Intel spun the FDIV problem, or the way Google and Facebook are spinning the NSA crisis, they would have said something like this: "It's a local problem in Chicago. Only a few of our customers have been effected. We are taking steps to make sure that there's a low probability of any other customers being similarly inconvenienced."
  • What a customer would think: "These guys are clueless. I'd better go with Bayer or Excedrin. I don't want to die just because I had a headache."
  • Instead, what Tylenol did, and it's something the tech companies would be well-advised to study, is to approach the problem the way their customers would. They immediately withdrew every bottle of Tylenol on dealer's shelves, everywhere. The first goal was to protect the people, then save the product. Which they did, after taking the hit, and on the way to becoming the leader in product safety. The product came back stronger than it was before.
  • Applied to this situation, it would have been smart for the companies to have prepared, by taking steps to blunt the negative effect on their users because of the government intrusion. Sure they had to comply, but did they have to leave the customers so vulnerable? Of course not. They couldn't warn us, that would have been illegal (unfortunately) but they could have made sure that more of the data resided in places that wouldn't be so convenient for the government to monitor. That's the equivalent of Johnson & Johnson putting tamper-proof packaging on Tylenol.