What to do when your ISP off-shores tech support

Martin Hannigan wrote:
> Hi Jay:
>
> Is there really anything wrong with sending first-level technical
> support offshore?

> Macs are macs, Windows is windows and mail is mail whether you're in
> Mumbai or Memphis. As long as the language skills are good and the
> people are well trained, it should be mostly irrelevant, IMHO.

In and of itself and setting aside patriotic/nationalistic issues,
probably not, assuming adequate technical and product knowledge and
language skills. I suppose that it would be possible that if it were
done well enough one wouldn't be able to tell.

Sure. Blaming off-shore tech support is pretty easy stuff, but the
reality is that the trouble is more along the line of appropriate
training.

For example, we maintain a Road Runner connection at the house, which
has been generally flawless over the years, with some notable exceptions.

I'll skip the DHCP-server-allocating-an-IP-address-from-a-netblk-recently-
vanished-from-the-global-routing-table story. Just *try* explaining that
to a tier 1... apparently my UNIX box was one of only a very few boxes
that hadn't re-DHCP'd in a year or two :slight_smile:

At one point, Road Runner introduced their "turbo" service here for a mere
$10/month more. Since it's nice to be able to download the occasional ISO
at high speed, and because it included a greater upstream speed, it was a
no-brainer. Worked great for maybe about a year. Then, suddenly, one day,
I began to see the modem crash anytime a largish amount of data was being
pushed through it. Spend time characterizing the problem. Spend time on
the phone. Get told the modem must be bad, get a replacement. You know
the runaround, so I'll omit the gory details. After a replacement modem
and the same problem, having spent several hours over the period of two
days on it, start raising enough noise through both the local and national
support services, talked to even the supposedly clueful people who were
puzzled, and one finally suggested calling some direct line to "a network
engineer."

Well, that actually turned out to be TWC Business. The guy was a bit
puzzled why I was calling *him*, but a brief explanation sufficed, and
within a minute or two he had the problem located ... the modem had been
only marginally sufficient for Turbo, and they had changed <something> on
the local cable that had broken it. Needed a *different* kind of modem.
Told me what to demand from the local cableco store, provided a ticket
number and everything.

Some discussion suggested that the RR people were highly script-oriented
and not necessarily capable of complicated problem solving. It appears
that the TWC Business tier 1 people actually have a fair amount of
technical training and clue, and resources to tap if that's not good
enough. Further, he was bright enough to let me know that they had a
"better than turbo" package available with a higher upstream speed, for
only a little more, that'd make me a business customer, so I'd never have
to deal with Road Runner again. Based on this one experience, we were
more than happy to sign an annual contract and pay just $10/mo more, and
have direct access to people who know what words like "DHCP" and "route"
actually mean.

I did ask, and all the local people are, in fact, local. It's a matter
of training and technical knowledge. None of them was really putting
together the fact that the modem was sketchy for the service class we
had.

My point is that you not only need the language skills and a good phone
connection, but also a reasonable process to deal with knowledgeable
people. I understand the need to provide scripted support, but there
should also be a reasonable path to determine that someone has an
exceptional problem and isn't being well-served by the script.

However, there is something about dealing with a local company that adds
value. People seem to care more about their community and neighbors
than a random, barely understandable voice on a G.729 8k codec at the
other end of a satellite link.

I have generally found dealing with most offshore tech support to be
very frustrating. The language issues are burdensome, some accents so
thick as to be barely understandable, and the lack of clue and scripted
menu-driven responses are obvious and usually of no value. I wouldn't
be calling if the problem could be solved by reading the documentation
and some judicious web searching.

That'll be the typical problem for this audience, yes.

There are some exceptions, including
Cisco TAC which is very good. I've talked to Cisco engineers in
Australia and Europe on occasion. I've had mixed results with Linksys
support, which I believe is in the Philippines.

Dealing with one offshore AT&T billing representative who was clearly a
non-English speaker was extremely painful. The latency and nonsense of
the person's responses suggested either some type of auto-translator or
satellite link, or both. The person wasn't capable of getting the hint
when I asked after several minutes of frustration what the "A" in "AT&T"
stood for, and in fact claimed to have no idea. I suspect that this
level of disservice may be deliberate so that people will pay bogus
charges on bills because the frustration level of disputing them is
intentionally high.

Yeah, ahaha. Like the "let's charge a late fee because we didn't promptly
process your payment" thing (another fun story).

Reminds me of the good old days of trying to contact somebody clueful at
some random network's NOC. Many of the same problems. However, the
operator community seems to have made good progress towards solving this
problem. So now I'm wondering why we're discussing this. :slight_smile:

... JG

Joe Greco wrote:

Sure. Blaming off-shore tech support is pretty easy stuff, but the
reality is that the trouble is more along the line of appropriate
training.

But, the reason that US-based $TELCO and $CABLECO use off-shore tech support is that they don't want to pay for the training and supervision to do it right in-house. The same person diagnosing your IP routing issues may indeed be asking, "Would you like fries with that?" thirty seconds later. [1] And, for purposes of, "Would you like fries with that?", off-shore is good enough that most customers can't tell, nor do they care. It may often be better than a newbie local ten feet from you. It's the ultimate scripted application, a literal menu. People expect half-duplex-low-fi audio when talking to a tin speaker buried inside of a plastic clown. :wink:

Some discussion suggested that the RR people were highly script-oriented
and not necessarily capable of complicated problem solving.

And they are afraid to admit (or don't realize) that they are not capable of complicated problem solving. They're following a script, just like the fast food order-takers. Or maybe they don't have the authority to escalate it to someone with clue, even if/when they do realize they're over their heads.

It appears
that the TWC Business tier 1 people actually have a fair amount of
technical training and clue, and resources to tap if that's not good
enough. Further, he was bright enough to let me know that they had a
"better than turbo" package available with a higher upstream speed, for
only a little more, that'd make me a business customer, so I'd never have
to deal with Road Runner again. Based on this one experience, we were
more than happy to sign an annual contract and pay just $10/mo more, and
have direct access to people who know what words like "DHCP" and "route"
actually mean.

I did ask, and all the local people are, in fact, local. It's a matter of training and technical knowledge. None of them was really putting together the fact that the modem was sketchy for the service class we
had.

So, regardless of geographic location, using scripted clueless order-takers without the ability to escalate for customer support is a bad thing. And, scripted clueless order-takers exist solely because they're cheap, not because they provide anything remotely resembling good service. Cheap, from a US-centric perspective, generally means offshore.

The interesting thing about your experience is that your service problems resulted in an up-sell, but only because you were persistent enough to fight through the system. Furthermore, it took a person with clue to do the up-sell. How many customers and up-sell opportunities does RR lose because of their decision to go with cheap, scripted, clueless off-shore support?

My point is that you not only need the language skills and a good phone
connection, but also a reasonable process to deal with knowledgeable people. I understand the need to provide scripted support, but there should also be a reasonable path to determine that someone has an exceptional problem and isn't being well-served by the script.

Precisely. Or for better service have reasonably clueful people at level 1 so that they can quickly and expeditiously deal with the easy problems that could be scripted.

The scripted part could (and often is) being done with IVR, no humans at all. But, please, if you do this, use DTMF menus and not that God-awful worthless "Tell-me" speech-guessing machine. And make sure that every menu has a "0-to-human-being" option.

[1] http://broncocommunications.com/

The problem with oursourced first level support is that they are totally
disconnected from real time operations and wouldn't be aware of problems
that network engineers are currently working on.

They have their scripts to answer the standard questions ("it tells me
to press the ANY key to continue, but there is no "ANY" key on my
keyboard"). But they are not trained nor do they have access to serious
diagnostic tools to help knowledgeable customers.

A good support person is someone who knows more about their own
network/product/serrvice than you do.

A bad support person is someone who only has access to the same
documents as end users (eg: the standard user guide) and is only of use
to clueless customers.

A good company would oursource entry level support to the lowest common
denominator, but make the script such that it is very easy for a
knowledgeable customer to get transfered to a "good" tech support.

Not always true. Our outsourced support in India were also our first layer of network troubleshooting, and they monitored everything related to the products they supported. They were almost always the first to call the engineers (in .us and .ca) to alert them of issues.

It's all about /what/ you hire them to do.....

...david

david raistrick wrote:

The problem with oursourced first level support is that they are totally
disconnected from real time operations and wouldn't be aware of problems
that network engineers are currently working on.

Not always true. Our outsourced support in India were also our first
layer of network troubleshooting, and they monitored everything related
to the products they supported. They were almost always the first to
call the engineers (in .us and .ca) to alert them of issues.

It's all about /what/ you hire them to do.....

Not only that. It also depends on the call center. I used to work for a
quite large call center, that would deal with anything from computer
support for vendors, cellphone support, cable-tv, cable-broadband, etc.
And just as an example for cellphones, the people on the floor had
access to internal systems of the telco's and where able to send
real-time commands to the switches.

When $TELCO decides to use this call center, it can sometimes take 2-3
years, before the calls end up in the call center. This is down to the
fact, that the call center has to implement structures with $TELCO that
will make a handover possible in the first place. Also stuff with enough
technical knowledge needed to be located within the agents or new staff
hired in.

Some customers had to be told, that it is impossible to do support for
them on the expectations, that they have, because their own internal
structures simply are a mess.

Outsouring and off-shoring is never the problem.

The problem is, and this was stated by the original poster, that the
lads off-shore he deals with have no clue and simply stick to the
script. No intention of looking what the real problem is. And that
problem lies not in the call center. It is the deal, that $TELCO struck
with $CALLCENTER and the procedures, that were put in place, that are
the problem.

Only solution: find a provider, who's support (off-shore or not) does
have a clue, has an escalation process and is willing to find a solution.

Kind regards,
Martin List-Petersen

That's a problem with a lot of internal first line teams too..

Offshore and outsource are different things, and when done right are irrelevant to the quality of service delivered. A willingness to offshore means you can deliver follow-the-sun NOC or support service, which can drive down delivery costs and health/safety risks for the organisation and drive up service quality by meaning that callers reach someone alert and awake ;-). Outsourcing offshore service makes it cheap and easy to do that.

Doing this well relies on building a process, and actually a different process for each network being supported, though I don't want to give away all the hints that I learned the hard way !

Andy

I did ask, and all the local people are, in fact, local. It's a
matter of training and technical knowledge. None of them was really
putting together the fact that the modem was sketchy for the service
class we had.

Yup -- I've had similar fun with Comcast. Once, I was seeing 15-20%
packet loss on the local loop and 90% (you read that correctly) packet
duplication. The advice I received translated to "clear your IE browser
cache". I demurred, and I was told that (a) generally, performance
problems were solvable that way, and (b) 15% packet loss was pretty
good. I escalated...

Then there was the time they upgraded the firmware in my cable
modem/NAT to a buggy release that didn't understand the activity timer
in the NAT table. Every 30 minutes, like clockwork, my ssh sessions
would die. I had to try to explain that to someone who didn't know how
to spell IP, let alone TCP.

Oh yes -- judging from their accents, everyone I spoke with was
American. In both cases, once I reached the clueful people, things
were resolved pretty quickly. (Well, not the packet duplication; that
took *weeks* to resolve, but once the packet loss problem was solved I
could at least get decent throughput.)

My point is that you not only need the language skills and a good
phone connection, but also a reasonable process to deal with
knowledgeable people. I understand the need to provide scripted
support, but there should also be a reasonable path to determine that
someone has an exceptional problem and isn't being well-served by the
script.

Customer records often include an optional data field that says things
about particular customers. I heard a story -- and I'll leave out the
names, since it's second- or third-hand and it does involve people and
companies most of us know -- that one very clueful person's record had
a note saying more or less "if you don't understand what he's saying,
he's right and you're wrong, and you should route his call immediately
to Tier N, where N is large"... But getting on that list is the hard
part.

    --Steve Bellovin, Steven M. Bellovin

I once had an @home rep insist that my connection was down because there was
ice in the lines. No matter how many times I told him it was 58 degrees
outside he stuck to his guns and insisted that was the problem.

Richey

How does one find such a provider? I'm unaware of any company
that lets potential customers test drive their $SERVICE call center
before purchase. Even if one did, how is a potential customer
supposed to evaluate the competence of said call center when
customer has no clue as to what problems may arise 5 years after
purchase of provider's service, whether said test drive provided
an accurate and appropriate solution, and whether said call center
quality will exist 5 years after purchase of the service.

matthew black
long beach, ca

Matthew Black wrote:

Matthew Black wrote:

The problem is, and this was stated by the original poster, that the
lads off-shore he deals with have no clue and simply stick to the
script. No intention of looking what the real problem is. And that
problem lies not in the call center. It is the deal, that $TELCO struck
with $CALLCENTER and the procedures, that were put in place, that are
the problem.

Only solution: find a provider, who's support (off-shore or not) does
have a clue, has an escalation process and is willing to find a solution.

How does one find such a provider? I'm unaware of any company
that lets potential customers test drive their $SERVICE call center
before purchase.

Ask others for their experience :), like for example here.

Even if one did, how is a potential customer
supposed to evaluate the competence of said call center when
customer has no clue as to what problems may arise 5 years after
purchase of provider's service, whether said test drive provided
an accurate and appropriate solution, and whether said call center
quality will exist 5 years after purchase of the service.

Well, if you're not any happy longer with the service, vote with your
feet again and find a better option. It's as easy as that.

Kind regards,
Martin List-Petersen

Of course, in much of the US, "vote with your feet" on residential ISP service might as well be as realistic advice as "pack up and move to a different city". [Perhaps not in the OP's case, though, if they are fortunate. Which it seems like they might be.]

- S

Skywing wrote:

Of course, in much of the US, "vote with your feet" on residential ISP service might as well be as realistic advice as "pack up and move to a different city". [Perhaps not in the OP's case, though, if they are fortunate. Which it seems like they might be.]

It isn't different here either :slight_smile:

Solution: if there is no alternative, it might be an idea to create one.

We had to do that here and works like a treat. You might find, that you
get more custom, that you wished for.

Kind regards,
Martin List-Petersen