Whilst away on vacation I received an email from EasyDNS to say that they had updated their interface. At first sight of this email I winced. Why? Not because I am a stickler for change – though no doubt some might agree – but because most of the time when there is an interface upgrade, there is no real “value added” when it comes to usability.
Any assessment and need to upgrade, update or change something should always be a value added proposition. What’s the point really in making it no easier, or harder, even, for your userbase to go about their daily tasks, right? So it leaves me even more mythed, when I go to login and review the new interface which seems to do nothing more than make the processes more complicated.
This is the second time in the last year that I have been faced with so-called web site upgrades, that leave me feeling nothing more than “rather meh” about the changes. So here are a few points I think are valid in tailoring the end result of a site or service upgrade:
Implicit User Feedback
The massive opportunity in doing an “interface upgrade” is that you have existing usage data, or at least you should, if you are doing any tracking at all. Knowing how people use your site, the process flows and such like, is a massive step in understanding how you can make their lives easier, reducing what might take 3 or 4 steps, and compressing them into one.
Usage stats, implicit in nature, are like a recommendation engine for any good UI developer. Briefing and informing them of the next best step. The primary question I would ask myself in the course of “upgrade” planning would be to understand what it is that users are doing, and what tools can be used to improve their processes, and no, AJAX requests and Web 2.0, are not one of them.
The real advantage of such stats at least means you are not making a complete stab in the dark when developing new functionality and layouts. If you ever make assumptions about use cases, when you have clear data, you are only doing your company a disservice.
Explicit User Feedback
Anyone will likely tell you that any web site or service should be defined by the users. The point about a wealth of usage stats underlines this very fact, but as clichéd as it is, so many sites and services online are designed and implemented without any regard for anything other than stakeholder’s intent (i.e. the business owners).
Obviously the larger your userbase the more fertile the ground to harvest from in terms of usage stats, but in the absence of this, the next best opportunity comes from directly questioning them. In the early stages this means identifying those who will provide useful and coherent feedback and tapping into that, from which you can take their concerns and issues and elucidate them in terms of new features (or NOT as the case may be).
Once you have implement this round of features, the next round of testing and feedback should involve more open beta testing, not launch or soft-launch as some developers deem fit to do. This process should bring about a final round of changes prior to a soft launch and/or further testing. Working with your userbase the entire way through the process is a definite key to success, as I have mentioned many times before, not least as a means to circumvent negativity that can come out of sudden and unexpected changes for users.
On-going Feedback
It seems obvious, but there is money to be made in listening to your users. Giving them the service they desire and need develops loyalty and new business. Providing support is one tool and avenue for on-going feedback, but more direct channels and requests for feedback are another means to solicit for free advise that can help improve your online offering. You might think you know your customers inside out, but they truly do know best, even if the feedback isn’t coherent or well collated, that’s is your job, and it is to your benefit to listen.
Above and beyond this explicit feedback over a long period of time, it is worth using A-B testing and tracking to ensure that your changes are creating the desired results and are not in fact impeding users int he course of their business on your site. This type of approach brings the view of updates full circle and provide a shining light on what can often seem like impossible to see dynamics in site use.
Making It Easier & Avoiding Pointless Changes
If your site or service has to use jargon, you are already heading down the wrong path. Nothing says more to me that a group of developers has hijacked a good ship than if you see a pointless footer tag, proclaiming “Generated in 0.215 second(s).” I am glad they got that one cleared up fast hahaha.
Product development and process management should be the driving forces behind changes, and these fact are underlined again and again in the points I have made. Not sure I can make ‘said things’ any clearer, and if it isn’t an improvement it shouldn’t be introduced. Even more so when resources and time are at a premium.
That being said, and to round things off nicely, if your are going to do nothing more than change the design and layout without improving form and function then you might as well not waste your money at all on any ‘upgrade’ to your site. Identifying hurdles and removing them is key to your success, and this should be at the forefront of your guiding light.
Don’t forget, that providing tools that allow your users accomplish their tasks at lightning speed and seeing them use it, not least register their appreciation of that fact, is very fulfilling, and it ultimately makes the job you do a lot more fulfilling.
Web-head & art collector, living in East London and huffing on the fumes of the planet since '78. Here are my thoughts.
Mark Jeftovic Jun 1, 2010
Hi Vincent,
I think your post outlines a great overall policy/roadmap for undertaking this type of endeavor.
If I may I clarify a few things, with an eye to explaining that this wasn’t something we did just for the sake of doing it.
The main reason we did the UI change was more out of necessity than anything else. I’ve always been the first to admit, that the old interface (while it worked) basically had gotten so old and clunky that we were handcuffed). Adding new features was excruciating and making any kind of code change usually resulted in something else “blowing up”.
The exercise was primarily to put us on a sound footing to be able to actually move things forward.
It has enabled us to add significant additions like IMAP hosted email, and a lot of the benefits that result from this are about to make themselves known soon (i.e. bulk DNS edits – never possible under the old system – are now around the corner).
(Instant activation of new domains, domain expiries via ical, enhanced geo-based login ACLs are a few others.)
What the new interface allows us to do is put even more control back into the hands of you, the member. So if there’s something you don’t like or a different way you want things to work (example: aggregating your renewal notices into a single monthly mailing or receiving them as they occur), you can do that, or you will be able to do that.
Believe me, if this was something we didn’t have to do, we wouldn’t have done it. I realize big changes are always difficult to manage, especially once your userbase runs into the tens of thousands like ours does.
Finally, because I have phobic aversion to “flash cuts”, we made the decision early into the process that we would NOT move everybody over automatically.
Anybody who doesn’t like the new interface or finds it inconvenient to port over at this time is free to use the old interface for the foreseeable future. (you can even leave some domains on the old one and move some over to the new one, if there is a feature you want access to for a specific domain)
Both platforms push data out to the same array of nameservers, so there is no lossage there (although major config changes like changing the IP of an external master nameserver on the new platform happens in a fraction of the time over the old one)
You’re right “this page generated in 1.34 seconds” is kind of useless information. I’ve asked the dev guys to turn that off.
Thanks again for your post, you obviously put a lot of thought into it and I’m sorry I never responded until now. It didn’t show up in my feedreader until today (that’s another story).
-mark
Vincent Roman Jun 1, 2010
Mark -
Many thanks for the extensive reply which I appreciate wholeheartedly. I have been using Easy DNS for some 10 years now, so I suppose I am a ‘diehard user’ and used to the old system, but at the same time understand you have to move forward with things. It wasn’t immediately clear to what extent the new system actually offered clear and obvious upgrade.
So long as this initial update provides a decent foundation for adding value later then definitely a worth while change, and it seems as if there are direct and immediate benefits from the get-go. Its hard from this perspective to know what is going on without full disclosure of the road map, and often in the past with other companies I have seen glossy changes that were tantamount to nothing more than fluff.
Your considered and solid approach underlies EasyDNS approach in all these things which has always been to provide a rock-solid service that operates in a bullet-proof fashion, and from reading your words alone it underlines this fact. You hardly want to jeopardize your business and that of your customers!
Anyhow, it will be interesting to see where things go, and I look forward to being an EasyDNS customer for another 10 years at least.
Vincent -