Hacker News Reader

Comment by mgkimsal | Hacker News Reader

"scaling support" almost always seems hand in hand with "dumbing down for lowest common denominator". By the time things have made it through "please restart your computer"-level stuff, maybe the only way to scale is to put real engineers on problems and "allow" them to actually talk to the people having the problems. This approach seems to be the only one companies seem hell-bent on never trying.

View on HN

I think the unfortunate truth is that "please restart your computer"-level stuff actually does solve whatever problem the customer is reporting enough that companies put value into that type of support.

If you're reading HN you probably don't need that level of support, but that's not to say there aren't customers who need their hand held to restart a computer.

> If you're reading HN you probably don't need that level of support, but that's not to say there aren't customers who need their hand held to restart a computer.

Part of the issue is that support is never allowed to push back on the customer. They don't get to say "I'm sorry, but you'll have to do a rudimentary computing course before I can proceed further". In a business environment it's all to easy to save on training costs in a department because they can push those costs on to support.

There really isn't a polite way to say "Do you have anyone competent in your office you could maybe ask first?"

This has been an ongoing problem for decades.

There's the (probably apocryphal) story of WordPerfect (huh, what's that?) support. It ends with:

SERVICE: "Tell them you're too fucking stupid to own a computer."

http://www.generationterrorists.com/quotes/misc/wordperfect_...

> Part of the issue is that support is never allowed to push back on the customer. They don't get to say "I'm sorry, but you'll have to do a rudimentary computing course before I can proceed further".

Well, no kidding, because who would pay for that?

Whether you're selling software or hard drugs, there'll always be customers who are just not worth keeping.

https://en.wikipedia.org/wiki/Angel_and_demon_customers

I'm sorry, but nontechnical customers are an overwhelming majority. If you can't figure out a way to serve them you aren't going to be very successful.

Because engineers are expensive, engineers that can talk to customers are even more expensive and decent engineers willing to do support are even more expensive.

That said, I do think software devs should be more involved with the customer for other reasons, far to often they are building software without having any idea what the customers workflow is.

Well, an entire support department that adds no value may indeed cost less, but losing your market because your product is defective (and you don't even know) can not be cheaper.

"expensive" can be relative. $100/hr but getting to the root of actual problems in, say, 2 days, and saving other customers from having the same issues in the future...

The same reason forcing devs to do ops on their own product justifies having devs do (some of) their own customer support. Otherwise they just toss any old crap over the fence and it becomes somebody elses' problem. Force them to solve those problems and they magically go away, cos devs hate support.

At one project we had that pushed over to QA. We didn't have to touch it. Report a problem form sent an email to JIRA allowing us to scrape the influx of 400 daily reports for potential issues. It was pretty helpful in diagnosing a Samsung Android issue where our app was killed for no apparent reason (they have their own additional resource manager that scores apps and kills them UNLESS the app has a persistent notification).

If possible, teams should always "fix it twice" from support requests - get the customer sorted ASAP, and then determine if it's possible to solve the problem fully (either extra support documentation, patch, longer term feature etc...).

Given that this is primarily an entrepreneurial site frequented by at least a few people who will build sizable companies in the future...

Maybe it's time that someone laid out a cost benefit analysis regarding offshoring support and presented a good alternative plan.

It's obvious to a lot of us techies, but there's a different perspective that we should be talking to if we want to see industry change for the better.

That's exactly what I thought the second I read this:

We would be needing further informations

It sounds uncannily as if Google has offshored service on Google Drive.

Or just hired people (perhaps on H-1B, perhaps not) for whom American English isn't their mother tongue, and who were hired for their technical rather than language skills.

Considering how badly Americans score on their English tests, I'd say quite a few Europeans speak American English better than Americans themselves.

is this ironic?

what's next? wondering if they have left you tube ui to chance?

I'm pretty sure they did. Along with some other services.

In Wrocław (Poland) we have a Google office. From water cooler talk I heard that it's mostly support.

From another source I heard about support calls where one side eventually figured out the accent and suggested moving from English to Polish ;).

Plural for information is valid in Polish and probably other Slavic languages (s. "informacja", p. "informacje"). A lot of Poles do a 1-1 translation when learning and it just sticks in their vocabulary (should work most of the time as Polish has a sizable subset of words with Germanic roots.