[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [cobalt-users] Cobalt Support

At 09:49 AM 1/27/00  Doug Forman wrote:


Ummmmmm...


mmmmm...
Your post cries out for an unbiased response. Which I can't really give, because I'm biased. I'm biased for several reasons. I currently lease a RaQ. I got it because I thought it'd be the "quick" way to do what I wanted to do, which was to replicate all the functions of the ISP I had just sold out of, one one system. (You ask how I could replicate selling T-1s, DSL and dialup all from one Cobalt Raq2? You do? Suffice it to say I can, but the answer is beyond the scope of this reply <smile>.)
My honeymoon was over the second day. Most things I need to do, I need to do "around" the user interface. Which means I could do them just as well on a cheap Intel box. Since the company I colocate charges me by the computer I colocate and NOT by the "u", I don't save colo money by using the RaQ. I'll always keep one, as it's part and parcel of selling T-1s, DSL, and dialup ("how?" you ask... again, not the topic of this reply). But in the future I'll use it for what it's good at, which plain and simple, is plain and simple hosting. The other stuff I do, I'll do on cheap Intel boxes.
I've had two experiences with Cobalt tech support as I recall, one with a helpful conclusion, one which led me to understand I knew as much about actually using the RaQ in the real world as Cobalt tech support, even though I'd had it less than a month at that point.
But I do understand the nature of the "tech support" beast, since I also own the site at <www.mailtraqna.com>, and we offer tech support for the products we sell there. 

That being my bias, here's my responses:


I think the Qube is a terrific example of design and engineering.


Absolutely.


  Because
it is a linux box at the core, the Qube is quite flexible and very
configurable for admins with basic -> intermediate linux skills.  I have a
Qube2 of my own, and have recommended Qubes and Raqs to several clients.
But only flexible if you have those skills, and aren't afraid of "breaking the warranty". Otherwise the Qubes and RaQs only do what they're sold to do, and only exactly the way the interface designers thought you'd want to do them. 


On the other hand...

My experience with "Cobalt Tech Support"  - that is, the "official" tech
support from Cobalt (not to be confused with this list!) - has been
abysmally poor.  I get "no answers" - "incorrect answers" - "inappropriate
answers" - and just plain "rude answers," but never yet have I received a
prompt, courteous and correct answer to *any* question which I have asked.


As I've said, my limited experience was slightly better.


>From the level of support which I have received, I suspect the support
"experts" answering email at Cobalt are part-time minimum-wage workers,
In this business you can't hire minimum wage workers. Even non-tech customer service people anywhere in California get $14 or so an hour. The customer support person has to be able to read in real-time at least <wry grin>. 


 with
basic training in the Cobalt GUI interface, and a couple-dozen
pre-programmed macro-keys on their keyboards for "canned answers" to most
questions asked.
Which is how tech support should be run. Everyone with the same question SHOULD get the same answer, no matter who answers the phone. Of course that requires a good customer support database, which requires years of experience. Which is why almost without exception when any company introduces a new product, technical support is abysmal for a while. It happened recently (and to some extent continues to happen) to Red Hat when they switched to telephone support. 


Questions not covered by the canned answers are either
answered incorrectly, or are not answered at ALL.
Questions not covered in the knowledgebase (their knowledgebase, not necessarily the same as the limited one they put on the website) should be escalated to second-level support, where someone SHOULD duplicate the problem, if possible, research the answer, reply to the problem, AND put it into the knowledgebase for the future. 

This last step is often where it all breaks down.


>From an article I wrote over a month ago (in which I copy-n-paste my emailed
questions and Cobalt Tech Support answers):  "...I suspect that xxxx's job
description is to dissuade as many customers as possible from bothering
Cobalt tech support."
Actually, it's quite similar. It's to answer as many calls as possible with a checkmark in the column "problem solved". If the tech feels he's answered the problem, he gets to checkmark that column. At the end of the day/week/month, he actually gets a bonus based on the number of checkmarks. I can't say for sure that Cobalt does it this way, but most companies do, and I'd be surprised if they don't. 


Cobalt reminds me of Apple Computer in the 80's - 'insanely great' products
and 'insanely poor' product support.
And for exactly the same reasons. The problems have NOT gone away. Have you tried Microsoft Tech Support lately? No? You don't even try anymore because you know how bad it is? Right. My point made. The problem starts the day the company gets so big that the owner can't do tech support for everyone him/herself anymore. And it never goes away. 


The primary reason we formed Macintosh
User Groups [MUG] in the mid-80's was to get the education and support Apple
failed to provide.  I co-founded a MUG in 1985; today it still has over
1,000 paying members - almost 15 years later.
Exactly. And that's the purpose of user groups. I've been president of a few myself (once got re-elected three years in a row before I realized I was working so hard no-one wanted to risk replacing me, so I stopped beating my brains out and lost the next election <smile>). Are you suggesting I start a Cobalt-users group and charge $15 or $20 per year (or even per month) per head? Okay, line up and pay me <smile>. 


  This [cobalt-users] mailing
list - and Luke Tymowski's QubeQuorner (http://weblogs.userland.com/qube/) -
serve as virtual-user-groups to get practical and "real" support for our
[Cobalt] product questions.  Where else would we go???
Where else should you go? Knowing how company tech support works, and knowing that none of the people manning the phones have ever had your problem themselves, and have at best only a disinterested interest in solving it, especially if it's complicated, and takes them away from solving ten other problems, why wouldn't you go to a mailing list, with actual users just like you? Sounds like a no-brainer to me. 


The obvious fact remains that, although this list is a form of "community"
for Cobalt product owners, if Cobalt was providing prompt, efficient and
accurate product support, 80% of the postings to this list would be
pointless and would disappear overnight.
I think what you've missed is that Cobalt provides this list as a very important part of providing prompt, efficient and accurate customer support. They're not dumb; they know we can give each other better support than they ever can. It's no doubt part of their business model. Certainly the Mailtraq mailing lists hosted by Mailtraq distributors around the world are part of my business model; if other Mailtraq distributors weren't already hosting them, I'd start some up. 


If Cobalt were my company, I would be very *proud* of my products, and
extremely *embarrassed* of my customer support.  But I suspect no-one is
embarrassed; I further suspect that this is all part of a pre-calculated
business plan...


Of course it is.  Why would you think otherwise.


push as much product (profit) as possible, and provide as
little support (expense) as possible.  Simple arithmetic.  Not rocket
science.  Not many other things, either...
The alternative is to provide more support (and to figure out the model to provide better support; I don't think any company ever has figured out a great customer support model), which is, as you say, expensive, and charge more to cover it.
That's not really a cost-effective way to go. I'm not sure of the numbers in Cobalt manufacturing, but historically, in America, most hard-goods products have always "cost" about one-fifth of their final selling price. And no, the difference is NOT all profit (a business-school educaiton is beyond the scope of this reply, too, I'm afraid <wry grin>). So for every dollar Cobalt increases their cost, their selling price goes up about $5. 

I'd rather get support from the helpful people on this list <smile>.


Perhaps your experience has been different.  Better?  I hope so.  I can't
remember reading the last posting to this list ranting about how "great"
Cobalt support is; perhaps I missed it?
There have been a few. There always are. But as a whole, the only part of the model that works IS the lists. If you can come up with a better model you can get rich, too. That's what Horatio Alger was all about <smile>. 

Jeff

--
Jeff Lasman, nobaloney.net
<jblists@xxxxxxxxxxxxx>
<www.nobaloney.net>, <www.mailtraqna.com>, <www.email-lists.com>