Dell Customer Service Strikes Again; DHL Strikes Out

This post is about customer service or rather the complete lack of customer service that I have continued to notice with many companies. I have had such a horrific experience over the last week that you can expect a whole series of posts about this issue about a Dell, DHL, Godaddy (and that's why I haven't posted in nearly a week). Last week I had a Dell 750 power supply drop dead. Of course it was the machine I had taken out of the cluster and doing mail service as well as some web hosting. (All of my current blogs EXCEPT this one were on the machine). Naturally I called the data center to request a re-boot, at first I assumed Apache had hung. The technician at the data center noticed that the machine didn't come back on nor did the panel lights come on. I had him quickly check wiring and everything seemed alright. So I called Dell. As you might imagine this is where things begin to go south.

After wading through the voice mail prompts, a pleasant Bangalore voice tells me I have dialed the wrong number for customer support. I check the web site again. Nope it's the right number, it's just the wrong number for support. So Dell's customer service number was wrong on the web site. After being transferred again and then again I actually speaking with a customer service representative.

Here's the first part of the conversation.

"Are you in front of the machine sir?"

Well no. This machine was purchased by me and then delivered to the data center by a Dell technician. I am not sure why Dell doesn't understand that.

Here's my response,

"No it was delivered to my data center in Los Angeles. I have had the local technician attempt a re-boot. He checked the cables and the panel lights have not come on."

"Could you have him give us a call from the machine?"

"Why?"

"Because we would like him to take the machine apart and reseat the cards."

"Well I had him check the wiring, the machine isn't booting and the panel lights aren't coming on. I think it's the power supply."

"Well we need to check the cards to be sure."

This is Dell hiding it's unwillingness to respond as a response. It's pretty passive aggressive. It's a rackmounted server in a data center, installed by a Dell technician. If the cards aren't properly seated in the machine, isn't the manufacturer's error? Why should I troubleshoot the problem? Because Dell doesn't want to respond to the contract I signed with them. Rackmounted machines

So I told them I would call them back after the technician had done the requested action. I waited ten minutes and called them back.

"No Dice. What now?"

"We will dispatch parts tomorrow. We have missed our shipping window for today."

"Ok great."

See I have a next day service contract with Dell which is what I thought would be sufficient response time. At the time the machine was part of a cluster so a single machine failing wouldn't be too big of a deal. My mistake when pulling it out to host blogs and mail - I would pay for it in lost revenue.

So I called Dell bright and early the next morning. This is the real kicker. Dell had shipped parts but they wouldn't arrive until tomorrow. This takes my next day response had makes it a two day response.

I called Dell pretty livid. In that conversation I found out the following.

1. Dell uses a just in time model for service. In short they stock NO parts in Los Angeles. That's right. One of the largest computer markets on the West Coast and Dell doesn't stock parts locally.

2. The shipping of the part is the triggering event for the service tech. NOT the original call to customer support. So I should have demanded the tech ship the parts that day. However I didn't know that they didn't stock power supplies for Dell Poweredge 750s locally nor that the shipping of the part was the triggering of the service technician. Dell's next day service response is actually next day after the part has shipped from Texas. So not next day service but rather

3. The shipping cutoff is 4:30 Central Time. This means my service call placed at 5:36 Mountain, 4:36 Pacific (where the machines are actually located) doesn't actually count. Doesn't it make sense that you should keep the same hours as your customers for a company of Dell's size? Shouldn't customers in California expect when you place a service call for next day response DURING normal business hours that you will get next day response.

After this little fiasco (the technician arrived and guess what? It was the power supply) I decided to up my service contract with Dell. So I called their service contracts division. This is also very revealing. Ordering an additional service contract will void my current contract with Dell. This contract which runs through 2007 would essentially be money thrown away. The new contract would run one year. It would actually be cheaper to buy a brand new machine with a new service contract.

I guess that highlights Dell's actual intention =-> sell more machines. Here's a better idea Dell. If you wanna sell more machines, how about a two pack with heartbeat running in the bios? You would sell two servers instead of one, get fewer customer services calls due to higher availablilty and your customers would be happier.

This post is getting too long so next time DHL strikes out.

Web 2.0 - Google Adsense's role

Well the buzz about Web 2.0 has gotten deafening at this point. In Silicon Valley hundreds of start-ups have Web 2.0 plastered repeatedly in their business plans. Sadly though most of these start ups will end up in the same bin with many of the dot coms. The process will take longer as the start-ups are much smaller, much more agile and more likely to become on going concerns but a lot less likely to become the home runs that VCs want. A lot of people have focused on AJAX, XML and open based standards as the factor in releasing the flood gates of innovation. I would argue another important (and obviously over looked) factor is the Google Adsense program.

Google Adsense allows a small group of developers to develop a cool web application, launch it and not worry about how to monetize the application. If this seems perfectly normal, let's go back to the mid nineties and remember how advertsing was sold on the web. If you were a start-up that had a media focus such as an online magazine (now we call them blogs or even just web sites) you needed to build into the site a spot for advertising. You then needed to put together an advertising sales teams who would sell ads for that slot. Congratulations! You have just added $1,000,000 in salary and benefit costs for a five person sales team to your previously small start-up.

This meant that you needed to chase VC or angel funding in order to generate revenue. Later you could outsource this ad sales function to someone like Double Click which was costly also. It also meant that a start-up spent a lot of time chasing deals. During the Dot Com heyday, advertsing revenue deals occurred left and right and were announced with much fanfare.

In many ways the chase for revenue skewed the business activity of the start-up. A ton of time was spent chasing revenue or VC dollars in order to buy ads to generate or chase more dollars. As a result too little time was spent on product development. I realize this a BIG generalization but why else didn't Geocities, AngelFire and the other free hosting providers evolve into blogs? I mean they were in the web publishing business. Why? Because all those providers were to busy cutting deals to jam more ads into their web application without actually paying attention to what their customers (ie the people setting up those free web pages) wanted.

In a start-up dollars are limited every dollar that is going into sales isn't going into product development. With the advent of Google Adsense, developers can begin generating revenue from the initial launch of the product from the very beginning. While this revenue might not cover salaries or the costs of hosting the application initially, they can certainly function as seed capital for developers. Furthermore as the Adsense network can be seamlessly added to site, it means that developers who are working on their own time or part time can spend more time with the application. There is no need to add expensive and largely unproductive sales staff. Google handles the entire process. In many regards Google Adsense IS AN OUTSOURCED AD SALES FUNCTION.

This allows a small team of developers to focus on product development, without the distraction of a sales team. And that's a key factor in the growth of web application start-ups.

Technorati Tags:
web 2
Google Adsense

SCO final puts up

It's taken 2.5 years and a ton of posturing in the press but SCO has finally submitted to the court a document purporting to list what copyrighted material that was lifted SCO Unix and placed into Linux. The five page document and another placed under seal which contains the full detail of the 217 seperate violations. I don't doubt that Linux and SCO Unix have code in common. Both are based in part on BSD - certain components such as the X-Windows system are provided under a BSD license which means that SCO undoubtedly has exactly the some code as Linux. This is taken from their own filing as their attorneys state,

"Some of these wrongful disclosures include areas such as an entire file management system; others are communications by IBM personnel working on Linux that resulted in enhancing Linux functionality by disclosing a method or concept from Unix technology," SCO said. "The numerosity and substantiality of the disclosures reflects the pervasive extent and sustained degree as to which IBM disclosed methods, concepts, and in many places, literal code, from Unix-derived technologies in order to enhance the ability of Linux to be used as a scalable and reliable operating system for business and as an alternative to proprietary Unix systems such as those licensed by SCO and others."

The key quote here is "from Unix derived" technologies. SCO is once again asserting a rather strange copyright claim, namely that any work inspired by AT&T Unix is theirs. This dubious legal theory seems quite strange considering these issues were resolved in the case. UNIX Systems Laboratories against Berkeley Software Design. In order for SCO to claim these copyrights, they are going to have to overturn this settlement. For a while that was their legal strategy but apparently their attempts have fallen on deaf ears. By December 22, 2005 IBM will have a response to the sealed motion. I suspect that this is going to go as well as the SCO VS DaimlerChrysler

Open Source CRM, Siebel, SalesForce

One of the initial successes in web based software was in sales force automation software. SFA was long dominated by the Siebel Systems, founded by Tom Siebel, a former Oracle executive. SFA shares the same problems that traditional ERP software shares; it's expensive, it's slow and it's a pain in the ass to set up and install. On top of that, sales force automation isn't a particularly difficult thing to do. I mean managing contacts and appointments with reminder notices? Not too hard as the paper version of this software (ie the organizer & calendar have been around for years). So this particular piece of ERP software was ready for webifying almost immediately. So whenSalesforce.com launched it took a huge chunk of the "on demand CRM" marketshare closely followed Rightnow Rechnologies. I am not sure of the purpose of artificially segmenting the market in this way into software and on demand market segments. These two market segments are certainly collapsing into each other.

Salesforce started out like many web based application companies by going after the middle market (companies 50-500 million in annual sales). This market typically has been underserved in the ERP software market cause it's too damn expensive with the typical seven figure depolyment costs. Netledger did this same approach with accounting software. I used their software until they wanted to charge me $399 a month - at which point I thought I would buy a copy of Quickbooks. So going after the middle market is a pretty good strategy. Typically these guys have customers and money to spend but not enough to really justify the big boys selling their software in a more scaled down shrink wrapped version.

So entering an underserved market and getting some customer wins are good way to prepare yourself to move into the Fortune 2000 - which everyone's business plan in the ERP space calls for. (Why anyone would want to sell to the Fortune 2000 is beyond me. They are notoriously stingy, demanding and the sales cycle is roughly 2 years. I once read a business plan which called for selling to the middle market to build credibility to sell the Fortune 500 (please note I use Fortune 500 & Fortune 2000 interchangably. Fortune 500 is US companies only and Fortune 2000 is the same type of companies world wide.) After selling several Fortune 500 accounts, they would use this credibility to move down market to THE MIDDLE MARKET. I pointed out that they would have better luck building credibility in the middle market and selling there since they weren't asking for enough money to actually complete a Fortune 500 sales cycle.

The problem of course for these software vendors is as I see it two fold. First off many of the functions they are replicating aren't particularly hard to do in software (contact lists? calendaring? contact management?) Web enabling these in software isn't that hard at this point. I am sure when Tom Siebel founded Siebel in 1993, CEO thought it was a pretty cool idea. Apply the same business management processes to ERP to sales and you turn all your sales droids into copies of your top performing. Hey maybe you can convert anyone into a effective sale person, thus eliminating the incredibly expensive commissions that your very best performers earn. Well your top performers remained your top performers and you weren't able to commoditize the position of salesperson.

This brings me to the real crux of this post. SFA & CRM are relatively easy to do. There is very little in the way of barriers to entry. There are a number of open source software packages which capture about 80% of the functionality of the Salesforce.com. SugarCRM is well on the way to being Siebel killer as 80% of the functionality is about all most customers need. Sure it might cost about the same to customize but you avoid the dreaded vendor lockin and the vendor upgrade treadmill. Open source is well on it's way to commodizing yet another software market segment.

Of course what's bad for Siebel is also bad for salesforce.com. With commodity software kicking butt, it's just a matter of time before someone takes the SugarCRM code and ports it to a web application to compete with salesforce. In fact that's exactly what is happening

Articles to Read
eWEEK Labs Review: SugarCRM's Sugar Professional 3.0
Open-Source Building Blocks Available

Marginal Competence - The unfortunate future of America

I have on the past occasion written about the pitfalls of outsourcing especially the particular exeperience I have had with Dell. Well after recent experience with another customer service snafu, it got me thinking about the outsourcing, the WalMart effect, under employment and the future of the American workers and American companies in general. The summary is simple, it's not a pretty happy future.

There are multiple trends vectoring toward a step decline in both the quality of the workforce in America and the product of that workforce, American goods. The first vector is the disalignment of the goals of the worker and the company he works for. Let's take an outsourcing example. Outsourcing 1st term customer service functions to firms in India and the Philippines creates a problem because the firms you are out sourcing to and your firm don't have the same economic interests. The outsource customer service firm has an interest in quickly completing calls so that their call numbers look good. They also have an interest in expanding their customer base of out sourced customer support firms.

Misalignment of goals is a common problem in customer support telecenters when they are run internally. Very often promoting people based on the number of calls they close, rather than a measure of customer satisfaction based on the call is always a bad idea. Yet this is a common practice in the industry. Outsourcing this function 1st tier customer support to another company in another country almost always decreases customer satisfaction. In fact I challenge anyone to provide one single example where customer satisfaction has improved as a result of outsourcing. Your goals as a company (satisfied customers) and their goals (high call answer count) are at odds. At best your outsourced firm is marginally competent at service.

Another vector contributing to decline of the American company that the goals of the worker and goals of the company are no longer as closely intertwined as they once where. Chronic underemployment is a problem throughout the economy. This leads to workers holding multiple jobs in order to make ends meet. Loyalty to the company and to the product that the company produces, is diminished. This means workers are increasing more loyal to themselves, or at least they must balance the multiple jobs pulling them in multiple directions. This most certainly has a negative effect on productivity.

Managers often tolerate the decreased quality because their management directive are to reduce costs. Hiring another worker who would work full time would incur costs. Replacing the under performing worker would also incur significant costs (training etc). Managers are often recognize that they cannot demand the same quality of a fully employed worker they can of a part time worker. They may think they might be able to but I would argue any worker worried about eating or making rent is going to be a lower quality worker.

This under employment is a corollary to the Wal Mart effect. Ruthlessly cutting costs means quality at some point begins to decline. Only so much inefficiency can be extracted from a supply chain before you begin skipping on quality of materials. When your only tool is a hammer, everything looks like a nail. Focusing only cost controls means that managers place less importance on other qualitative factors. Anyone that shopped at WalMart would concur, the quality of goods is simply lower. (We have a WalMart here in Las Vegas. I have stopped certain types of goods there such as electronics).

Living with marginal competence of work means that anyone who can beat your cost of goods and labor. This creates a race to the bottom for wages, and raw materials. One good test for this would be the mean failure rate for consumer electronics. Are electronics more or less reliable as a whole now or 10 years ago? I haven't done that research yet. I will do it and report back.

Technorati Tags:
outsourcing
Walmart
Walmart Effect

Google & Sun - The Slow Bleed

A lot of bloggers have been writing about the implications of the recent Sun Microsystems and Google alliance. I believe this alliance is still the formative stages so it doesn't mean too much right now. I think Google's approach is going to be avoid going directly at Microsoft Office. Why? Because it's the previous generation of technology and not wholely appropriate for the web as it's so document/desktop focused.

As a recent article article in The Register quotes Brin as saying,

“I don’t really think that the thing is to take a previous generation of technology and port them directly,” said Brin. However distributed thin web applications allowed you to do “new and better things than the Office package and more.”

Well clearly Mr. Brin realizes the value of a web application is the fact it's on the web. This provides quite a bit of value by itself. As Office is so desktop centric and pretty document centric, porting Open Office to the web would be silly. Why? Because too much developer time would be spent web enabling the very boring and none too useful document centric features. The next generation of collaborative applications will be web focused and web centric which creates the value in and of itself.

Clearly Mr. Brin understands Google's place in the software universe. If Google suddenly buys Jotspot you will know they are quite serious about de-throning Office by changing the paradigm of collaborative work. I have long argued that the thing that is going to de-throne Office won't be a poorly made clone but rather a new way of working. Given Office's desktop/document production focus it's unlikely to make a radical shift. The MS Office product manager more than likely talks to MS Office customers about features. He doesn't talk to people who might be using an entirely different paradigm of work flow.

How does the slow bleed work? Well by capturing the customers of the neqw subversive work flow paradigm, they slowly over years bleed customers away from Office toward the new collaborative paradigm. This will take a few years (I would guess till 2018 or so) but eventually the Office market will evenutally come to resemble the mainframe market - slow, steady and a lot smaller.