Qliktech Global Partner Mobile App Contest – The Runners & Riders: 1 Thoroughbred, a Few Also Rans and Lots of Old Nags
As some of you may be aware Qliktech has been running a Global Partner Mobile App Contest with the top 5 entries winning an invite to the upcoming Qonnections 2012 event in Miami and yesterday they released the ‘Top 20’ to the Partner Community for voting. For those with Partner Portal access you can view the entries here: http://www.qlikview.com/us/landing/partner-app-vote
I was hoping that I could use this post to showcase some great innovations in how to develop Qlikview apps for mobile devices (and there are one or two which I’ll outline below) but I am quite frankly appalled at the utter laziness and ‘wtf-iness’ of the majority of these apps – these are from Qliktech Partners (in many cases ‘Elite’ Partners) who should be at the very least following basic – and I mean basic – dashboard design principals. I’m also not going to pass comment on any of the technical capabilities of the dashboards as I can’t gain access to the scripts of expressions. Finally to those who may say; “why didn’t you enter if you think you know it all” – well a) I don’t know it all and b) I’m not a Qliktech Partner and the contest was only open to Partners, I am in the process of building a demo which I’ll release if at all possible for wider scrutiny.
Firstly a few points; I’m not going to single out specific Partners and developers for critisism as that would be un-professional, I’m not going to comment too much on general aesthetics as I know that can be a matter of opinion; I’ll try to stick to clear cut well documented design issues. Also I have no relationship and have never worked directly for any of the Partners mentioned.
I’m going to start with the bad stuff to get it out of the way as they really are crimes against dashboarding (and the Qlikview name):
I know I said I wasn’t going to name specific Partners but there are 2 I have to mention as they are both Global names and the worst offenders for not even getting the basics right: Cap Gemini and Tata Consulting; you should be doing better (much, much better); their entries I’m sorry to say look like they’ve been thrown together in minutes with no thought.
1. Incorrect Chart Choices:
Why oh why is a Line Chart (via a Combo Chart) being used here? Basic charting tells you that you can’t use a Line Chart for a non-continuous axis; a Line is to show transition from one X-axis value to the next eg: days or months, what relationship does ‘Hang Seng Index’ have with ‘IBRX 50’? – None; so why join them together? Instead the line should be replaced either by Datapoints or simply Bars. This is just one example from across the 20.
2. Heavy Use of Pie Charts
I’m not going to wade into the debate on whether Pie Charts are good or not but I do feel they have their place in very isolated instances but they’re the most commonly used chart type in these dashboards; you can’t tell me in all cases the Pie Chart was preferable to a Bar Chart. Amoungst the Pies there were one or two horrors including this one:
3. Screen Resolutions not Being Correct
This is surely the first thing you sort (and to be fair the majority have got this right), you know this is going to viewed on an iPad, you know the screen size is 1024×768 so why not develop for that screen size? This is just one of many examples of lazy developing here.
4. “3D is Good in Cinemas so it Must be Good in Qlikview”
Any dashboard design book will tell you; ‘don’t use 3D charts’ as they distort and obfusticate the data contained within so why are so many partners using them in what should surely be their flagship apps? Here are just 2 of the worst examples:
For me it’s clear cut; there’s no time I can think of when a 3D representation of an inherently 2D chart is a good idea; so why does Qlikview allow users to do it? (Tableau – I know I’m banging on about it alot at the moment – doesn’t even give users the option and it’s all the better for it.)
5 “Adapting for Touch Means Spreading Everything Out”:
There are a few examples where entries have taken into consideration that users of these apps are going to be using touch to interact with these apps; such as wider scroll bars, buttons with actions etc; I’ll outline some in the ‘good’ section. In the main the stratergy seems to have been; ‘Let’s just spread everything out’ – come on; it’s about more than that.
6. Just Plain Wrong Charts
If you’re going to use a chart then you have to actually think about what you’re showing; I see so many cases where developers have created a chart clearly without thinking how the user is going to read it, for me this is an example of the worst mistake:
What do those Gauges show us? – there’s no value on the axis; the user can’t tell what (presumably) 106,490 is a proportion of; useless. Gauges are like Pie Charts (of limited use and over used in these apps) in that they show what something is as a proportion of a whole; I always use the example of a church roof collection the ‘Thermometer’ is useless unless you know what the target is:
Even this guy get’s it right (he’s annotated the axis) and I doubt he works for a dashboarding company.
7. Other Areas
i: Inconsitent colours running through apps – it’s easy so get it right.
ii: Bad colour choices; if your bar is a similar colour to the background you’re going to cause problems.
iii: Not tidying up; more than one entry still had the ‘Send to Excel’ and ‘Print’ icons on their objects; it’s on an iPad; you can’t send anything to Excel!
iv: General sloppy, lazy design; obscured axis, labels overlapping, objects not lined up, background images the same colour as text the list goes on.
I don’t feel any of these critisisms cover anything complex; it’s all basic, elementary stuff that I’m sure the designers in question are aware of but for whatever reason don’t bother to implement.
On to the good stuff:
The by a ‘Country Mile’ Winner of this Race:
Snowview by Vince Wolbers & Rob Nales of Victa BV in The Netherlands. Their entry is so far ahead of the competition it’s a no brainer for the win (and a few thousand $ from Qliktech), I absolutely love it – I’m not all about negativity. It was the only entry that truly adapted it’s app for use on an iPad and did it in a thoughtful, meaningful and innovative way – well done chaps.
There are a couple of really good innovations in play here over and above the use of good design, effort and thought; there are images as buttons designed for touch usage, great use of images to engage the user etc; it really is what a Qlikview iPad App should be.
Othere than SnowView there were a couple of others worthy of mention:
Project Brokers main entry (the second isn’t that great) almost gets there with the design and uses some good bespke menu items:
These are quite simple to create using bespoke images whosr display is controlled by variables which in turn are controlled by Text Objects with Actions ‘Set Variable’ Actions assigned to them, it’s a technique I often use. (One tip to Project Brokers; make the menu display buttons ‘minimize’ the other menus as you currently end up with nasty overlaps.)
Other than that the only other notable entry was from Differentia Consulting who again appear to have put a good amount of effort into their entry again by using several bespoke elements.
So why are so many of these ‘Top’ 20 entries so appaling, these are after all from Partners who should be at the pinnacle of what Qlikview can do. I don’t think there are any definitive answers but I think the following are at play:
1. Partners are in a high pressure environment; if they aren’t on site billing for time then they aren’t making money so it’s a case of get it done and move on. That in my opinion is no excuse; surely you should invest time in creating really good samples and templates (as far as they’ll get you) for use in the Sales & SiB process – that’s what I tried to do when I worked for a Partner.
2. Most Partner Developers are ‘Techies’ and don’t know much about design. There’s nothing wrong at all with being a Techie (some of my best friends are Techies) but what is wrong it putting these people in charge of designing full applications including the look and feel; something they unsurpisingly have no experience of. You wouldn’t for example let the person who designed a cars engine design the bodywork would you: so why do the equivalent in Qlikview? Partners really need to invest in some talent that knows what they’re doing when it comes to layout.
3. Laziness. It sounds harsh but I can’t think of any other word for it; the failures in these apps aren’t complex, each and every developer surely knows the basic; so why don’t they put it into practice?
All in all a bad showing, I do commend Qliktech for running this competition and 3 entries are on the whole good but the rest just aren’t good enough for what Partners should be offereing; plus the fact that the Top 5 entries will get a free trip to Miami which means 2 bad apps are going to win their creators a free jaunt – not good.
Needless to say I’ve voted for SnowView and I implore you do the same to ensure the cash prize (and other goodies) go to the deserved winners. Take a look at the voting site (if you can) and you should at least be able to get a few useful insights.
Qliktech: Why do you on the one hand promote good dashboard design (‘don’t use 3D’, ‘maintain colours etc’ through various presentations and channels) and then go and actively promoted apps that go in completely the opposite direction? – you need to be more of a Gate Keeper; if it’s not up to scratch; don’t associate with it, you’re just entrenching bad practice.
As always; I hope you find this useful.
All the best,
***UPDATE***; Victa & SnowView have been declared the Winner of the 2012 Mobile App of the Year contest at Qonnections in Miami – Congratulations. Hopefully the standard will be better next year.