Russ Casenhiser

Open Letter to Google Engineering: Please find a better testing process

David Szetela from Search Engine Watch posted an open letter to Google Engineering asking them to please slow down a little.  The basic assertion is that Google Engineers are moving so quickly that they are not taking the time to thoroughly test new code.  And as a result, bugs are making it into live code.  His proposed solution is to slow down and test more rigorously.

But here's the thing, Google: you're already waaaay out ahead – for example, your Google AdWords campaign management software is two generations or more ahead of Yahoo and Microsoft. So you can afford to slow down a bit and test more rigorously before even minor software releases. You're starting to get a reputation for developing buggy software – one that's not deserved for the most part – and now is the time to nip the problem in the bud before the perception starts to snowball.

While I agree with the goal of producing code with fewer bugs, I think the proposed solution is misguided.  Google has successfully created a culture of velocity and execution that has been tremendously successful. Slowing down would be a significant mistake.   In a recent New York Times interview Google's CEO talked about velocity,

VELOCITY does, indeed, matter, and Google deploys it to great effect. Conventional software is typically built, tested and shipped in two- or three-year product cycles. Inside Google, Mr. Schmidt says, there are no two-year plans. Its product road maps look ahead only four or five months at most. And, Mr. Schmidt says, the only plans “anybody believes in go through the end of this quarter.”

Speed is a key advantage, and not something that should be sacrificed lightly.  If Google is in fact way ahead of Yahoo and Microsoft, and I believe they are, then more power to them.  In fact I hope they keep screaming along.   I'd rather see Microsoft and Yahoo fall so far behind that they have no choice but scrap their old models and reinvent themselves into leaner, faster companies.  Besides, slowing down doesn't guarantee better quality.  We can point to endless examples of slowly developed buggy code from Microsoft.

A better challenge to Google engineering would be to create a testing process that can keep up with their current velocity.  If they've been able to reinvent application developement, why not testing as well?

 

The Ascension of Strangers

A few weeks before Christmas my mother and I were chatting about ideas for holiday gift giving and she semi-casually mentioned a shiny gizmo she'd like to have.  The next day I was killing honey-dos at Costco and walked right by the exact shiny gizmo we'd discussed the night before. [Cue Angelic Choir Music] Quick inspection revealed it did exactly what I wanted, made by Sony, reasonable price and simple enough for mom.  Sold!  One more thing off the list...   

As I put it in the cart, the questions started in my head...  Is this really the best one?  I wonder how it's been reviewed?  Is this really a good price? 

With an internet connection I could resolve these doubts in about 60 seconds, but I was pushing a sled through Costco and no relief for the nagging questions could be found.  Reluctantly, I returned the shiny Sony gizmo back to its home on the shelf, and moved on.

Now I am not an indecisive shopper, in fact quite the opposite, and I wondered about where this sales process broke down.   Eventually it dawned on me that I've come to trust the opinions of often unqualified strangers more than the promise of a brand.   This is where the tried and true sales models like AIDA really start to break down.  The product got my Attention, I was clearly Interested and had Desire, but in the end didn't take Action. 

Why mention a century old sales model? Because I still see it used and referred to all the time.  Most recently it popped up in Call to Action, an otherwise great book on persuasion architecture for the web.  The model's had tremendous longevity because it does work in many cases and has proven effective for focusing and improving sales results.  But it fails to explain why I hesitated and ultimately decided against purchasing that shiny Sony gizmo.

For me, this potential purchase was missing Confirmation from a trusted source.  My trusted sources take many forms; user reviews of movies, user reviews of restaurants on Yelp, knowledgeable friends, user reviews on Amazon and some expert reviews.  Most of these are people I don't know, probably will never meet and likely have little "expertise" in the area reviewed.  Yet I've come to trust their opinion because more times than not, I've had good results.

A brand or an expert used to provide the confirmation and trust that's often necessary to complete a purchase, but no longer for me.   I'll take the opinions of strangers over the stamp of Sony and be better for it.  Interestingly, about 15 minutes later, I walked into the wine section of Costco where they prominently display the Wine Spectator "score" of nearly every wine they sell.  I picked up a few bottles of a 93 pt. cab without a second thought.

So I have to ask, does it make sense for Costco to label and score hundreds of cases of wine, yet not provide the same confirming information on the few dozen electronic items that cost 10 times as much?  Is there something you're forgetting to confirm that could make the difference between sold and passed by?

Got Pine

Lessons from a tree farm. 

We've had an artificial Christmas tree for as long as I can remember.  Our children, 10 and 13, have never had a live tree in their house, until now.  Being a live tree noob, I was apprehensive about the process of acquiring a tree, not just any tree mind you, but the first tree of our children's lives.  I ran through the list of usual tree selling suspects, home depot, various tents around the neighborhood, and then saw a sign for Orange County Christmas tree nirvana... Peltzer Pines 7 miles that way.

Now granted, we're talking Orange County here, so pine trees are not exactly growing on every corner.  Orange County is a desert turned green by copious amounts of water delivered from Northern California every day.  Peltzer Pines is a small farm that has been growing Christmas trees for 40 years.  And with that experience, they have completely nailed the process of acquiring a Christmas tree.  There are lessons to be learned ...

1. Simple works.  There are no rfid tags, bar code scanners or even credit card machines on the tree farm.  There's a two part tag on every tree.  You wander the farm until you find the tree that calls your name, you rip off part of the tag, pay for it, then have a handy lumberjack chop down your tree.  Said lumberjack will even clean it up, fit it for your stand, hoist in on your vehicle of choice and tie it down.  Simple, easy, immediately understandable, and can be operating by children.  They do one thing, and they do that one thing extremely well.
2. Directly relevant upsells never feel pushed. The only options for dropping more money were delivery and a tree stand.  The tree stand was designed Mr. Peltzer and like everything else we experienced, is simple, good quality and just plain works.  The upsells were easily accessible throughout the process and never pushed;  the lumberjack and cashier simply said "We can deliver if want, and here's a stand that works better than what you have now."
3. A quality product never has to shout we're the best.  I've seen the trees in parking lot tents and piled in front of Home Depot.  30 seconds into wandering the farm, it was immediately obvious that these trees were better; that cutting my tree down and taking it home same day was better; that this process was better.  Their signs don't shout we're number 1, or we're better than Joe's tree shack on the corner.  Live trees.  Cut yours and take it home the same day.  It's just better.  No need to shout it.
4. Help should be there when you want it, but not in your face.  Half the fun was chasing our kids through the trees and yelling, "Look at this one, no wait, how about at this one, but then there's this one over here."  There was no hovering sales staff, but  help was there when you wanted it.  Mr. Peltzer was greeting people as they arrived and answering questions, the lumberjacks had lots of tree care tips and the whole staff knew way more about Christmas trees than I did.  Friendly, knowledgeable, accessible.  Perfect.

So a tip of the hat to Peltzer Pines.  They got it right.  Oh, and you can save $4 if you're really interested.