Updated: 7/3/06; 12:26:01 PM.
Ed Foster's Radio Weblog
        

Friday, June 02, 2006

Support professionals will tell you that the vast majority of software support incidents are caused by user error or user failure to read the documentation. So does that mean software publishers are justified in charging for any support that actually requires individual attention? One of my recent stories provoked an interesting discussion in that regard.

A number of readers argued that if support isn't paid for by the user everyone, including customers who solve the issue on their own, will wind up paying higher prices. "Someone, somewhere has to pay for product support," wrote one reader. "Let's face it, nine times out of ten it's not a problem with the product but some user setup issue. The fact that your reader eventually got it going by himself indicates to me this is one of those times -- surely it's reasonable to charge for handholding?"

But other readers thought that the original situation involving Nuance (formerly known as ScanSoft) makes the opposite point. "I, too, have no problem paying extra for extra support, but my experiences with Nuance have been horrible," a different reader wrote. "Their survey was much, much longer than anything I had ever had to fill in for any other company in the 20 plus years I have in PC software support. I HAD been a big fan of ScanSoft software, but in the last few years, it has been big, buggy, and slow."

Others suggested that the increasing lack of documentation and customer access to knowledge base information undercuts the rationale for paid support. "I agree that paid support makes sense," wrote another reader. "Someone has to pay for it, and 90% of the time it's the user's configuration or setup causing the problem. My objection is to the lack of an available knowledgebase. The companies that make you register to look at the knowledgebase or worse -- downloading patches and updates -- are just as bad. I don't want to spend 10 minutes providing the marketing department with information before I can troubleshoot a product I paid for. Whoever the current owner of EZ-CD Creator lost a customer on that one. So my solution is to make the knowledgebase publicly available and easily accessible. Give us access to the same scripts you have your level one support monkeys reading -- if I still need your help to troubleshoot my computer, I will pay for it or return the product."

If you've paid for a product, however, others feel strongly that the software publisher does owe you some minimal form of support. "Out of a duty to customers, all software for individual use --enterprise use software should have a separate set of rules -- should have at least a minimum limited-free tech support period, wrote another reader. "Thirty days would probably be enough in most cases, but 90 days ought to cover everything. After that, I don't have an issue with charging for support."

After all, what if the product simply doesn't work? "I object to paying support fees for software I have purchased," wrote another reader. "If I purchase a program, I expect it to work properly. If it doesn't work properly, then it is the fault of the program, not of myself, even if it is an incompatibility with something else. I'm absolutely livid about those companies that charge for bug reports, especially in the cases where the user isn't even expecting a response, perhaps because they've discovered a workaround themselves. That's just rampant stupidity and money-grubbing."

In that respect, it's difficult to consider the question of paid support in total separation from the issue of software quality we've been discussing. "That's called the 'software lifecycle,'" wrote another reader about paying support for buggy software. "You get an unusable version 1.0, crummy version 2.0, a marginally usable version 3.0, and a fully functional but fairly buggy 4.0, and then a decent 4.1 or 4.2. Then you get a bloated 5.0, a slow and crufty 6.0, a broken 7.0, an unusable 8.0, and if the gods have mercy that's where it stops. Eventually someone else makes version 1.0 of new software with the same sort of functionality, and the cycle starts again."

So what do you think? Should customers always pay to cover the cost of support, pay only if the company can prove the answers weren't readily available in documentation or knowledge base on their website, pay after some minimum number of days or support incidents, or pay only if the software itself is free? Answer the poll on my website, post your comments, and/or write me directly at Foster@gripe2ed.com.

Read and post comments about this story and respond to the poll here.


11:28:25 AM  

© Copyright 2006 Ed Foster.
 
June 2006
Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30  
May   Jul


Click here to visit the Radio UserLand website.

Subscribe to "Ed Foster's Radio Weblog" in Radio UserLand.

Click to see the XML version of this web page.

Click here to send an email to the editor of this weblog.