The MoPR Blog

Magicians have a credo: Always keep the magic a secret. The logic is simple. Once the secret of a trick is revealed, the magic is gone and the trick is no longer compelling.

Sure, there are magic acts that trade on revealing these secrets. Penn & Teller for example do nothing but reveal the secrets behind some of magic’s most popular illusions. But when they do, they always end up producing an even bigger, more impressive trick in the end. That’s rare.

Magic is an apt public relations analogy for what a great many technology companies do.

Many technology companies have their own magic that makes their products and services compelling. Unlike magicians, however, who perform in front of an audience that seeks to be amazed, technology companies perform for an audience that simply wants the “trick” to work. In fact, the best services are often the ones that are so simple in concept and predictable in usage that users of the technology are assured that it works every single time. A customer has a need, a company has a solution. How that solution works is often irrelevant.

At a former company we often talked about “masking the ugly plumbing” to describe the simplicity of a mobile IT security service built on a complex technology backend. It was enough for reporters, industry analysts and indeed our customers to know that we had patented technology that produced a magically simple solution. In fact, if we were asked what our secret was, the answer we provided was something like, “Does it matter? What we do is effective.” And that answer was typically acceptable.

That’s what makes the demonstration by voice-to-text technology provider SpinVox so strange. They went out of their way to violate the magician’s credo — with no new trick up their sleeve — and showcase exactly how their magic works. The result was a PR disaster. I have to ask them (rhetorically) why?

Here’s the back story: SpinVox has been lately described as “embattled” because of claims that they have burned through $100 million in venture funding to build a service that cannot possibly scale. SpinVox offers an automated system for transcribing voice messages into text. Their claim is that 100 percent of the messages are automated to some degree. The counter-claim is that rather than relying on their technology to handle transcriptions, the vast majority of their messages are transcribed by human operators. As SpinVox suffered in public for their financial situation (it is reported that they are paying some employees in stock and some older vendors not at all), they staged a PR event to reverse their perception troubles. If their intent was to stop themselves from circling the bowl, what they achieved was pulling the handle another time to increase the speed of the flush.

I have seen SpinVox at many trade shows. They typically have elaborate booths and hand out expensive giveaways of Ken-like (as in Barbie and) dolls with the SpinVox cube for a head in exchange for a demonstration. I’ve done this demo a few times and (only took one doll) and it usually worked. The widespread rumor was always that there was a human being doing the transcription. I never cared. I could see a value for this service. Value means it can be sold. Sold means revenue and revenue means it can pay people behind the curtain to do the transcription. Supply and demand. Even if a machine did all the transcription there is still a cost. So the question SpinVox really needed to answer was how can it scale, not how does it work.

Instead SpinVox invited a group of technology reporters to its “plush” UK headquarters to receive a complete demonstration and have a thorough Q&A about its technology. In the end what was demonstrated is that virtually all messages wind up being transcribed by a human being, not by their technology. Although it was reported that when the patented aspect of SpinVox’s technology did in fact determine their system was unable to handle the transcription itself, it handed the message off to a human operator, thus perhaps validating that 100 percent of their messages are automated to some degree.

The press was not kind:

“All of the messages supplied by our small group of visitors tripped through to a human operator. The event was unnecessary and humiliating for all concerned. SpinVox shouldn’t have had to lift its skirts; we didn’t need to be there,” reported Andrew Orlowski in The Register.

“Here’s where it got ugly. From observing the ‘tenzing’ (the SpinVox technology) process in action, it was clear to us that the system had failed to pick up a single word in the message correctly,” Milo Yiannopolous reported in TechCrunch.

And Ben Smith of The Really Mobile Project enumerated the key points of the demonstration thus:

1. The presentation and demo made no attempt to address the most serious allegations against SpinVox.
2. The demo, even under controlled circumstances, failed to demonstrate anything more than very basic automated transcription. Most messages went to the human transcriber, whole.
3. The system we saw showed a ‘build date’ of 29th July so may not be one that has transcribed many (if any) real customers’ messages.

“The DNA of SpinVox has been based on a lie. I am not sensationalising, I am pointing to a number of truths picked up by others,” wrote Bena Robers in GoMo News under the headline “Ouch SpinVox more fatally stupid flaws.”

There was more bad press in silicon.com, The Guardian, paidContent and telecoms.com, among others.

Whew!

So what can we learn from all of this?

For one, it has been my experience that reporters and bloggers don’t like to write bad reviews or critical articles, that is unless the product, service or company is truly deserving. For one thing, who has that kind of time? With so many other products, services and companies on which to report, why would the media want to focus on mediocrity or failure? They don’t. If they go out of their way to open the door to such a negative press opportunity, company executives cannot complain if the media walk inside.

Second, battle-test your demo before you send the invitations. Put it through “worst case” scenarios. The surname of one the media people invited by SpinVox was “Yiannopolous” — run tests on that name before Milo enters the room. Reportedly one of the words the automated SpinVox system failed to transcribe was “SpinVox.” Ouch. If the system can’t transcribe “SpinVox,” then “Yiannopolous” may likewise be problematic.

Third, and perhaps most importantly, know what the media are really going to care about. In every one of the half dozen or more articles I read on this SpinVox PR debacle the same question was rhetorically posed: “but can it scale?” This question was truly at the heart of the matter. SpinVox is a company facing financial not technological scrutiny. It did not have to reveal its magic. It merely had to prove longevity. It never answered the scalability question. In the end, all SpinVox did was expose the secret of its illusion making the service far less compelling. As Jack Schofield concluded in his post in The Guardian:

“I quite like the idea of a service where you leave a voice message and someone types it out. But haven’t people being doing that for a few decades now? So what’s all the fuss about? Either I’m missing something, or the world is even crazier than I thought…”

These are lessons that SpinVox CIO Rob Wheatley was forced, it seems, to learn in the heat of battle during his demonstration. As Orlowski reported:

“Towards the end, after prolonged badgering on the machine:human question, an exasperated Wheatley asked, “Does it matter? What we do is effective, and we’ve got carriers.

“Indeed, if SpinVox had taken this approach from the start – ‘we use humans, so what?’ – we wouldn’t all be where we are now.”

Epilogue

To make matters worse, SpinVox was forced to learn another PR lesson, this new one about transparency. In the aftermath of all this bad press, the company was accused of “astroturfing,” a practice in which a company employee posing as an unaffilliated reader posts positive comments. As PR Week reports, this is part of SpinVox’s “ongoing PR crisis.”

The company did just receive £15 million in new funding. Hopefully the company can recover and reverse all this bad press through its future success.