Sunday, December 6. 2009Providing "Support" Is An Interesting Activity.....John Linton .....not least because it is almost impossible to define what "support" actually is or how good it is. A long time ago in a company far, far away and after listening to various experienced people talk about what "support" actually meant in the context of supplying broadband services I suggested that a broadband suppliers obligation to its customers in defining where "support" ended was in the provision of an at designated speed service to the customer's 'network boundary' which, for the purposes of that definition would mean an ability for the supplier to ping the customer's modem at an agreed acceptable speed of return. This was in the days when most customers had to install a 'software client' to connect their crude ADSL modems to their ISP's service and that client was opsys dependent and modem hardware dependent and was more difficult by a large margin than today's plug and play devices so it was decided that actually helping the customer configure their interface (hardware and often software) would move the boundary point back to getting the customer to be able to ping a nominated site at an agreed return speed would be required. I have never changed my mind that a steady "link light" was all that was required and that 'touching' the customer's modem (let alone anything the 'other side' of the modem) in any way or assisting in the configuration of the modem itself in any way was not within any realistic definition of providing support for a broadband service. In practice I never saw these guidelines adhered to as almost all support engineers have an engineer's innate characteristic of demonstrating their engineering knowledge by going far beyond such an arbitrary (if totally reasonable) boundary and most customers know how to 'plead piteously' for additional "support"....particularly female customers. As modem hardware became more sophisticated and 'software' add ons became things of the broadband stone age this stricture should have become easier for support engineers to adhere to but it didn't prove to be the case. Over the years, and that is now well past nine, I have given up on attempting to 'enforce' that very reasonable definition of what constitutes a broadband supplier's responsibility of providing support to a customer and I don't see anyone else in Exetel making any effort to restrict "support" to only ensuring the customer is connected at the correct speed. So, over the years, I have slowly changed my personal, and in Exetel's case, the company's view of what is included in support for the broadband services we provide. I have talked with a number of broadband providers, both in Australia and in the UK, about what support 'boundaries' they employ in providing telephone support to their customers and have found no consensus in any aspect of doing that. There don't appear to be any actual boundaries set down in black and white beyond everyone I spoke with having objectives for both call answer times, call abandon times and quantities and for call talk times. I haven't found any company that measures customer satisfaction with the experience of their calls; not because they don't regard it as important but because they haven't found any meaningful way of measuring it (beyond the mechanistic aspects of wait times, resolution times etc). As a person brought up in business life with quotas and targets it is an inevitability that I will think of any aspect of Exetel's performance in terms of black and white numerical measurements. When we transferred the residential support functions of Exetel to Sri Lanka we realised that their would be a knowledge loss and other issues to contend with that we believed could be ameliorated in the medium term by deploying larger 'numbers' of people and that would mean we could offer longer support hours and shorter wait times in the shorter term and the knowledge loss would be addressed over time. To a large extent that has proven to be the case: 1) Support on week days has double form 8 hours a day to sixteen hours a day 2) Support is now avaialble 8 hours a day on weekends and public holidays 3) Call wait times have reduced by 40% As the number of support engineers has continued to increase the average number of 'months on the job knowledge' per engineer has obviously not increased (a mathematical impossibility in a rapidly growing 'head count') but it is now beginning to increase as the number of support engineers that have been with Exetel in Colombo for 12 months or more to increase. So one of the issues we addressed while I was in Colombo for the November review was how we would set the expectations for support in 2010. The simple to measure call wait times, number of abandoned calls etc are not difficult to plan for and we will continue to aim at reducing those times/amounts along the path we put in place at the start of operations. How to measure the improvement (assuming there is any) in the quality/effectiveness of the support provided still eludes me; as does 'controlling' the breadth of the support provided on topics that we should not be addressing. So, any ideas on how such a thing can be done would be welcomed.
Trackbacks
Trackback specific URI for this entry
No Trackbacks
Comments
Display comments as
(Linear | Threaded)
This is where the old dichotomy between qualitative and quantitative measurement comes into play.
Many years ago, when I was involved in a technical support environment, it was always hard to establish meaningful qualitative measurements, because of the subjectivity of both parties involved, who see both the problem and the resolutions offered from different points of view. Regular surveys of customers who have logged faults, with a number of questions that have graded answers will provide some measures, but this can become unwieldy very quickly, unless properly automated as part of the fault management system. Additionally you will only receive meaningful responses from a subset of callers, which needs to be large enough that it is statistically significant. With respect to the "boundary line" - as you say, it's relatively easy to set down a criteria, but a lot harder to overcome human nature in going that little bit further. Collecting information as to what the fault resolution actually was, can build up a picture of what level of assistance is being provided. This could be implemented by having a range of "tick the box" categories for the support personnel to select when the fault is resolved, but in my experience, "Other" seems to always be the major resolution . The other side of the coin, is setting customer expectations as to where the boundary point is. I would assume in broadband supply, you have two major types of faults to deal with - not working at all, and not working to expected performance. Unfortunately, from what I can see in various online fora etc, is that most customers don't seem to be able to differentiate between the two. The difficulty in setting customer's expectation to a set level or boundary point, is the vast scale of experience / knowledge that customers have, ranging from those that get the modem/router out of the box, plug it in, and hope it works, to those who have sophisticated home networks. All of them "buy" a connection from you, and believe that it's up to you to make the connection work. Much as I dislike Telstra, taking a leaf out of their book, their process of applying a fee if the problem is not within their area of control (ie, inside the customer's premises, etc) seems to ensure that most of their customers make an effort to isolate where the problem may be. I don't know if there is a magic bullet to sorting out the dilemma you're facing, but I'd suggest that collecting information on specific call resolutions, etc, as mentioned above, may be able to identify where work needs to be done. This work may be as simple as closer management oversight of the support personnel, to start with. Comment (1)
Hi Klaas,
Thank you for your detailed response to my request for assistance - it is very valuable. Have a very enjoyable Christmas. John Comments (2)
Do your support engineers have on hand a list of your agents to refer customers to?
Perhaps agent support could be more widely utilised (paid direct to the agent by the customer) if the problem does not appear to be within the support boundaries. I had a new HSPA customer who I talked through her installation over the phone and all went smoothly & she was able to connect. I received a call back later in the night to say that her pc was now saying that her usb device (modem) was not recognised. strangely her camera was also not being recognised when plugged in.] She rang support, who thought that the modem may have been damaged somehow, and who advised her to uninstall & reinstall the modem. Problem was, she could re-install due to the unrecognised problem. There is a Windows XP fault that can cause mass storage devices to stop being recognised. The fix involves editing the registry, which should only be done by experienced technicians. I advised her to bring the pc to me, 1/2 later, the modem was working fine & she didn't mind paying me to have the XP problem resolved. In cases like this, agents can provide another level of support. Comment (1)
Hi Anne,
It's a sensible idea and we will consider it. It was raised on the agent forum (different but the same) ad it is being considered. Comments (2)
Following up each support incident with an opportunity for the customer to anonymously provide feedback on their experience is critical, in my opinion. In terms of what questions to ask, refer to my reply to Steve Waddington in the comments section of his recent blog entry on the subject:
http://steve.blogs.exetel.com.au/index.php?/archives/221-Customer-Feedback-Forms.html#comments Comment (1)
|
Calendar
QuicksearchArchivesCategoriesBlog AdministrationExternal PHP Application |