It’s the Putting Right That Counts – An Open Letter to Net24

An open letter to Net24 – sorry if it makes you wince but I really hope (for your sake) that you think about the issues I raised..

Dear Net24

So I’m pretty pissed off – so sorry if this email gets a bit aggro.

We host our website with you – it’s on a VPS, something I’m not overly keen on (you see despite being a technology commentator, I’m not, and don’t want to be, a sys admin) so you sell it as an “empty container” and don’t support it. That’s fine when it works, when it doesn’t though, it sucks.

Over the time that we’ve had the VPS, there have been quite a few outages. Annoyingly they’ve been out of office hours and, despite a totally misleading name, you don’t actually have staff around 24/7. I can call an on-call engineer who only wants to hear if there is a new outage – otherwise he’ll just run through his schedule fixing things that go down.

Yesterday we sent out one of our regular newsletters to around 5000 recipients. Given Murphy’s law it was obvious that the VPS would go down just as we were sending a campaign. Luckily this time it went down an hour or two before and I checked the site before hitting “send”.

I called the engineer who told me that you’d had a hardware outage and he was going through a list of affected servers and copying them to a new node one at a time and that he’d get to us “sometime soon”. I told him that wasn’t good enough and he agreed to bump us up the list – 15 or so minutes or later we were finally up. Yay! I hit send and watched around 5000 emails hit cyberspace…

And yup – ten minutes later a buddy tweeted me to say that our site was down. It seemed that in moving the VPS as a whole (remember when you said that when moving a VPS you do it in its entirety and nothing can go wrong?) somehow a MySQL database got screwy and, whatever the propeller head reason, we lost our site.

So I gave you a call and at first you did the “no, it can’t have been us. you need to get your developer to go through the logs to find the problem”. I, of course, jumped up and down and got some extra help and you fixed the problem (thanks!) and we were back up and running. I sent an email suggesting that maybe, just maybe, it might be an idea to inform people proactively when their servers were down and not just leave them to find it out for themselves – that’s something you’re thinking about I guess.

So then this morning… our customer service rep told us that she’d not been getting emails from the e-commerce app sitting on our server. This is kind of important as the emails carry half of the credit card detail (we’re concerned about our customers security so half goes via email and half is through the app control panel) and we couldn’t process orders. I called up and was told that “moving a VPS you do it in its entirety and nothing can go wrong” and that VPS are unsupported and we should get our developer to go through the logs to find out what the problem was.

What? So I have to pay a developer to find out exactly what the problem is when you caused it through a hardware failure? Whatever happened to the consumer guarantees act? Fitness for purpose? So after jumping up and down (good exercise I guess, but bad for my blood pressure) I pretty much forced you to look into it yourselves – eventually you did and you resolved the issue, but like I said – only after some serious prodding from me. At the time of posting this letter our email is working, you’ve agreed that “there was an issue” (something to do with file permissions) and we’re up and running.

So what gives guys? I could have spent a few hundred dollars with a developer fixing your fault. I could have had a website that was out for several hours with no proactive advice. I could have lost significant sales while our email was mysteriously disappearing. Twice in 24 hours you swore black and blue that the separate issues we experienced were not in any way caused by you – both times your contention was proved wrong.

You see it’s all about putting things right.. you’ve got a Twitter account now, that’s great. But without a culture that is actually about making sure customers expectations are at a very minimum met, and, ideally, exceeded you’re just chirping into a great big void.

I hope some things change.

Ben

12 Comments
  • no-one actually cares or takes any responsibility for problems basically…

  • @Simone – well I’m hoping they’ll see this post and think about what it means for their business. For every one person who publicly complains, there’s a bunch who just move their business elsewhere. That’s a recipe for disaster for any business….

  • Ben, refer to my previous comment a while ago:

    http://diversity.net.nz/diary-of-a-site-owner-chapter-1-breathe-deep-breathe-deep/2008/12/12/?nomo=false#comment-15834

    ?Why aren’t you eating your own dogfood?

    Cheers

    Ben

  • Is there a better alternative? Sounds like they are not abiding by any contract T’s and C’s. Meanwhile nice post – I like the idea of helping companies get better, provided they listen.

  • @ben – an Amazon EC2 server *is* a VPS, it just has a different management/provisioning process. If Cactus Climbing self-hosted their webserver I could understand your comment, but getting a company like Net24 to host a VPS is no different to using Amazon’s EC2. A locally hosted VPS has a better response time and is probably cheaper than EC2.

  • (I meant “@Ben Reid” by the way, not “@Ben Kepes”, but I’m sure you all realised that.)

  • @Ben (Reid) – Oh but we are totally eating our own dog food. The dog food that says “A business should stick to it’s core”. You see Cactus is in the business of making the best outdoor equipment in the world – NOT of being a great sysadmin. If your comment had been couched in terms of “eat your own dog food and go fully managed hosting” then maybe you’d have had an argument – as it stands I totally disagree…

  • @Lance – Part of the problem is that it is a VPS. Generally ISP’s love this as they can get away with putting all the onus of service onto the customer. And in a case when the fault is still on their side, they can shrug their shoulders and walk away….

    Look out for more “open letters” int he future….

  • @Stuart – Agreed, the issue here is lack of service. If anything a move to EC2 would be far worse. Bear in mind we’re mass market, non-tech customers… yup – that’s the reality out there in the real world!

  • I disagree… A VPS is nothing but a hosted server – just virtual and living on a large machine with other virtual servers.

    Many hosting providers will give you the (paid) option of a totally managed environment – no different than me using my own big box having a contract saying ICONZ will do all the management – patches, upgrades, etc (I just don’t have this because I can do it).

    But you get the idea – somewhere (sorry, can’t be bothered checking their site, I have other stuff to do, such as comment here) they should offer a complete solution as in “your VPS, our team”. Don’t they?

  • @Mauricio – the problem is ISPs tend to think of a VPS as totally unsupported.

    In their defence they do say “Net24 will provide support for the VPS environment, but will not provide support for applications on the VPS. Because a VPS functions the same as a real dedicated server, you will require server administration skills to successfully manage a VPS” but more often than not the thing causing the need for support is their fault…

  • Providing support for clients like yourself is what I do for a living.

    Often clients have a unrealistic expectation of what support will be given to a particular product, even when the terms are written in a signed contract. In the heat of an issue the hosting provider is always at fault in the eyes of the client when more often than not its closer to home than expected.
    Companies who have staff who know what they are doing seldom have any issues with sorting out problems, however many budget businesses turn to VPSs as a cheap alternative and expect the hosting provider to manage and fix all problems, both platform and application issues.

    Granted if the issue is on the side of the hosting provider then they should be either proving that its not or be busy working on the fix and not telling end users to troll logs and sort it out.

    Also your article doesn’t mention any past history with net24, if for example a client was frequently the cause of issues then the hosting provider would be less inclined to prioritise the troubleshooting. I think there is a kids story about a boy and a wolf that comes to mind.

    Good luck with the VPS, over time they grow on you :)

Leave a Reply