If You Ask It, Pass It

It’s not fair — all the United bashing going on around here — but the recent post titled “Don’t Make Me Scream” over at SvN struck a chord, especially since designing and writing speech applications for large enterprises like United is what we do at Versay.

I couldn’t agree more with Matt’s rant. Many speech IVR applications out there are terrible, not just United. Too often I find myself walking down a street on my mobile phone trying to get information, only to be greeted by a cheerful voice asking way too many questions and getting seriously confused by the sounds of trucks and cars and other random background noise. Knowing how these systems are designed makes me an even more demanding user. I know they can do better and it is frustrating that best practices are so infrequently used.

A particular pet peeve is the situation where I navigate through an application, patiently providing things like account information, reservation information, and the like. . . and then when I am transferred to an agent. . . yep. . . I am asked the same questions all over again. Unacceptable. As a client of ours once remarked (when describing our shared golden rule about the hand-off between automated and live customer service): “If you ask it, pass it”. Granted it’s not always the easiest feature to implement, but for my money it is essential.

So, Matt, I hear you loud and clear and I can assure you that we are working with our clients every day to improve their UIs and to re-think high-quality, yet cost-effective customer service.

800-GOOG-411 (466-4411)

Update (4/11/07): Two Tellme executives head for Y!.

It’s been a Google kind of week here. Google Labs has just released their much-anticipated and long-speculated speech-enabled free 411 service. This comes on the heels of Tellme’s recent Business Search beta, offered behind their 800-555-TELL (8355) service.

Some early and incomplete observations:

  • Tellme’s Business Search is very. . . well. . . Tellme, including the usual navigational audio clicks and swooshes to mark list items and returning to the beginning, as well as transitional music and service announcements. These things when not used judiciously tend to slow down the call unnecessarily and also betray Tellme’s entertainment industry pedigree.
  • Google’s service has tell-tale old school Nuance VUI elements for things like error handling, contextual help, and offering options like “details” just before the call is transferred.
  • Unlike Tellme, the Google app appears to be music-free and lacks earcons, although they do play a humorous “thinking” sound no doubt mocking the typical cue for system processing. This is in keeping with Google’s overall stripped down approach to interface design.
  • Tellme entertains and then optionally texts you a listing but does not connect the call (yet). Google connects you (and optionally texts if that is preferred). Which do you think is the better experience?
  • Google has an interesting back-off strategy and allows callers to enter requests using their dial pad. This is a nice feature for noisy environments or simply when speech recognition is having a tough go of it.
  • Google’s text-to-speech sounds signficantly better to these ears, although this improvement creates another interesting challenge. The synthesized speech at times is almost too close but not close enough to recorded human speech, evoking an uncanny valley experience. It’s creepy at times.

Overall, I’d say the services do a fairly good job of locating what you want as long as you steer clear of uncommon requests. For example, Tellme sent me to a post office in response to my “postcards” request (not really what I wanted) while Google was sure I needed to talk to the coast guard. (To its credit, Google kinda got it right the second time around, recognizing the request properly but then offering to connect me to the Pleasure Chest in Chicago, citing it as a. . . um, “related listing.”)

I was much more successful asking for “stationary” on 800-GOOG-411. I was quickly connected with The Paper Source on Armitage Avenue. I’m sorry to report this request was equally perplexing for Tellme though. Granted, I am being a bit unfair, and by no means am I arguing that these systems don’t work well. In most cases, things went smoothly. But with these limit cases, it was interesting to see how each service reacted. It’s one of the most important aspects of the work we do at Versay — designing applications to get callers back on track when things go wrong. These factors are amplified dramatically when you are working with the large number of options required by automated directory assistance.

As Om Malik points out, Google’s entry into this space, while no great surprise (do I sense a bit of a yawn, Om?), certainly spells trouble for pure-play providers like 800-FREE-411 (373-3411) and 800-411-SAVE (411-7283). And no doubt Yahoo is just around the corner. On the other hand, it is a definitive vote of confidence in the viability of large-scale speech application deployments and one hopes a sign of further innovation to come.

The Next Google Revisited

Last year, I posted about the “next Google” phenomenon sweeping the nation, specifically comparisons between 37signals and the search giant. This year, the landscape hasn’t changed all that much. Companies are still hotly competing to get a piece of the search market, start-ups and established technology firms (e.g., Microsoft and Yahoo) alike.

37signals are still going strong and have released their CRM-lite application, newly named Highrise. I personally think it is their best release yet and could potentially absorb/cannibalize some of their other offerings. While I get why they prefer to keep project management separate from contact management, there is enough overlap between services so that some core data elements should be shared. Just as Apple engineers have resisted the all-in-one approach of Microsoft Outlook, instead offering siloed apps like Address Book, Mail and iCal, they nonetheless enable sharing of elements common between these applications. It’s proven to be a successful strategy in balancing usability and functionality.

Interestingly, if one were to ignore the numerous and obvious differences between the two companies, it could be argued that Google and 37signals have actually moved closer to one another in terms of service offering over the past year. With the Google Apps launch, they too want to provide the tools to help you manage your business in a hosted, web-based model. And why not? Google already provides services for email (Gmail), calendaring (Google Calendar), and instant messaging and VoIP (Google Talk) not to mention “Microsoft Office killers” Docs & Spreadsheets and Page Creator. The Google Apps offering rolls these services up into a complete package in three flavors: small business, enterprise and education. And from a technical perspective, most importantly both companies have embraced APIs, enabling third-parties to deliver “mash-ups” of theirs and other services. This is almost a given in today’s 2.0 world and arguably one of the key drivers behind the success of hosted services.

Google’s move signals a couple of things.

First, hosted application services for business and not just the consumer space have arrived in a very big way and definitely threaten Microsoft’s desktop dominance. If nothing else, both the live.com initiative and recent Tellme acquisition confirm this. This is not to say that hurdles don’t still exist. While bandwidth and availability concerns have largely been overcome by large capital investments in infrastructure, other factors such as data portability, privacy policies, and security still remain stubborn obstacles for hosted solutions. With Google in particular, given its transparent goal and not-so-transparent methods to index the world, there is still a strong distrust of the company’s motivations and a nagging fear around the vulnerability of hosted (implicitly shared) corporate data.

Second, as Miquel Helft points out in his January 1st New York Times article, in its effort to go after communication and collaboration, be it for small to medium-sized businesses or large enterprises, Google risks losing ground in the search arena. No doubt just as AltaVista and Webcrawler were once default, now abandoned search options, so too Google’s reign is constantly threatened by innovation just a click and a bookmark away. That is, of course, unless the search business has changed so dramatically in the past seven years so that new barriers to entry will thwart a dramatic shift in market share.

Given the current climate, one wonders how steep of a climb a start-up like Powerset (one of the latest entrants in the “next Google” sweepstakes) will have, or if search is in fact the fastest way to claim the title for keeps, as last year’s front-runner, Wikipedia, seems to believe.

Chiming in on MacTel

After the initial wow factor of Apple’s announcement of Boot Camp and support for Windows on their new Intel-based hardware, I’m now a little less enamored with the idea. The two things that I haven’t missed about life with Windows XP is the less than stellar plug-and-play experience with peripherals and the inevitable sluggishness and disk churn caused by Windows caching and file system fragmentation. I went through two hard drives on my previous WinTel laptop and I’m convinced (evidence be damned) it’s in part due to the stress the OS puts on the hard drive. Now, I know with Boot Camp everything is neatly sandboxed and partitioned, but I can’t help thinking how will Windows behave with the physical hardware? Will drives die faster? And, back to plug-and-play, will I have to cross my fingers every time I plug in a USB thingamajig?

Some folks I’ve talked to think that when Windows users get their hands on a Mac, their introduction to Mac OS X will result in a ready-made Switch campaign. Maybe. Maybe not. But I wonder if rather than a rising tide lifting all boats to meet the excellence of OS X, we may instead see Mac hardware flake out just as badly as the Dells and HPs out there. I mean, let’s face it, we aren’t talking about the most rugged platform (my original Mac 128K will forever be known as the MacMelt, thanks to a burned out power supply).

The question boils down to: what kind of company does Apple want to be? Hardware? Software? Entertainment? Platform? Applications? Lifestyle? And how strategic or reactionary is the Boot Camp move? Is it a clever way to accelerate migration from G4/G5? A way to quiet the gamers? And what is the fate of Microsoft software for OS X like Office and Remote Desktop? Time will tell.

The Next Google

If there is any doubt that 37signals, purveyors of fine application services such as Basecamp and Backpack, has jumped the shark, Information Week has sealed the deal with a recent fluff piece headline: “Is 37signals the new Google?”

Don’t get me wrong — I am a long-standing supporter of the 37signals crew since we first approached them to design a site back in 1999, and a true believer when it comes to their unwavering dedication to clarity in design, and “less is more” approach to business management. I also can’t help but be impressed with their evolution over the years, growing from a usability and user experience-oriented design shop to a blog-powered, seminar-driven marketing machine (in the best sense of the word). And, while I don’t “live” in their software as some of their biggest fans do, I would agree that they are among a handful of companies designing very good software today. I eagerly await their entry into the hosted CRM market. SalesForce could use a serious wake-up call.

But the next Google? Back to the headline.

Jason Fried, master of ceremonies at 37signals, shared news of the article with his Signal v Noise blog readers, generating a typically active discussion (41 comments as of this post).

Over at Del.icio.us, 76 people (and counting) also find it link-worthy. Tags include: 37signals, google, web2.0, and ideas.

Which got me thinking: what other companies have been suggested as the next big thing in the age of Google?

A Google search for the phrase “the next Google” produces 90,300 results (though Google cuts you off at 440 claiming the rest are similar). Among the top hits, you will find:

and . . .

It will be interesting to see where these companies, Google included, rank a year from now, 5 years from now and beyond. I wish Jason and the 37signals team the best.

Structured Blogging

Fall 2005 was a blur. So much so that I decided to take the blog portion of the site down. Too much going on. But now, with a new year ahead, and in response to several requests, I’ve brought the blog back online and couldn’t resist the opportunity to restructure content a bit.

As I’ve noted before, I use Movable Type to manage content. I’m a big fan and have convinced others to use it for more than just blogging. Even with WordPress growing in popularity, I find MT’s balance of structure and flexibility to be a good fit. For example, I use a single MT weblog to manage different types of content, from blog entries like this to the various static photo series to my photoblog. It’s probably not the most elegant way to go about things, but by customizing MT’s standard templates combined with a handful of plug-ins, the site dynamically responds to different media types and the structures they require, in both human and machine-readable formats (e.g, the feed for this blog).

So it was with considerable curiosity and excitement that I found the Structured Blogging site by way of Paul Kedrosky via Dick Hardt. I have to agree with Mr. Hardt — it just makes sense for this type of innovation, especially the ability to produce microcontent-specific feeds. By baking in much of the customization I’ve hand-rolled, MT will be that much more powerful out of the box.

Best wishes to everyone for 2006!

Digital Self-Storage

In his recent Slate article, “Self-Storage Nation”, one-time Baffler regular, Tom Vanderbilt, describes America’s healthy appetite for space to store stuff.

As Vanderbilt hints, the recent growth in self-storage is evidence of a larger trend that stretches well beyond those ubiquitous sheds one finds scattered along the outskirts of town. Companies like Container Store, publications like Real Simple and books like David Allen’s Getting Things Done all seek to answer the growing demand to organize, simplify, and store the things that make up this modern life.

And it doesn’t just end with thing things. The same could be said online: Google’s Gmail offers 2GBs of free storage (easily hacked to store and serve files rather than email) and various services like Xdrive and sephoneSAFE offer the safe and convenient storage of one’s growing digital media files (e.g., .mp3, .flac, .jpg, .raw). Even the software I use for this website, Movable Type, can be thought of as a content management tool, a means to produce, organize and store stuff.

In most cases, this content is saved on hard drives in web hosting facilities that we will never see, never visit, never really give more than a passing thought. In fact, these places tend to resemble those impersonal sheds on the edge of town. Is it just a matter of time before Public Storage goes digital?


I’m a micropatron. I am someone who has contributed hard cash to Jason Kottke’s recent foray into full-time blogging. I’ve never met Jason, never even exchanged a single email save my recent well wishes for success, but I’ve been a long-time reader (and not so infrequent poacher) of his site. Like many successful blogs, Jason’s voice is strong throughout — he’s subjective, daringly honest and tireless. He’s also perceptive and provides a well-manicured digest of ideas, memes, and the random what-have-yous circulating the web at any given moment.

My “patronage” is not just about supporting this kind of writing. It’s also about supporting web writing as a professional endeavor outside of an advertising-based model, and rescuing blogging from mainstream media’s ill-informed dismissals of amateurism. In other words, it’s a vote of confidence.

I wonder what it’s like to be in Jason’s shoes right now — to witness an outpouring of support from readers who have been relatively anonymous or unknown until now. What an incredible leap of faith it must have been to trust us, and an equally incredible validation.