Monday, 1 May 2017

Challenges in travel technology - a holiday makers perspective

I love travelling, not because of going to new places - although that is great! - but because I enjoy the journey. Mainly because I'm a travel and systems thinking nerd. My latest holiday was a chance to look at airlines and airports in a different way. Scheduled flights to hubs for business are one thing, package holidays to small airports proved another thing entirely. In fact, comparing Gatwick airport to Heraklion it is a good example of this quote
The future is already here – it's just not evenly distributed. 
William Gibson
It got me thinking about two major challenges to objectives of the IATA fast travel program being the norm across the whole industry.


The first is customer culture. Self-service common use technology is still in its infancy. Previously there has been a consistent expectation for example of what travel means, and how desks work, this is now being replaced with different tech and varying security rules. For example, even a large hub like Amsrerdam can have its issues

So far every "self-service" experience I've had at an airport has been slightly different. The only constant factor are there being a number of staff on hand, who are there to tell you how to use it. 

There is also the difference in business vs leisure/package travellers. For example, what use are home printed bag tags if your holiday rep is picking bags up and checking them in. So, all you have to do is go to security gate (best case) or if you have children you need to tag push chairs etc

Business travellers are also likely to have more idea about the self-service capabilities of the airlines and airports that they use most often. So giving leisure or infrequent travellers awareness of the airport facilities to prepare them can only be a good thing. 

But still, the experience and expectations vary greatly between airports. For example, a smaller airport like Heraklion where not all the check-in counters even take the luggage, here you have to get the bag tag from a counter and then take to the x-ray machine. Here it is not a simple case of installing a few kiosks and some software, there is a much larger infrastructure and business process improvement project to consider. Which brings me onto the next challenge.


The second issue is with technology roll-out. I don't see a major engineering problem needing solving. Additionally the problem isn't consumers, after all most "industrial" tech stacks a behind the consumer experience from finance and insurance to health and travel. It isn't the airlines, who although they have legacy systems largely in place there are providers that can deliver modern communications. It's the airports. Or more specifically the spoke and small leisure destination airports. IATA acknowledge this in their risk assessment:

Business case: Airports may be reluctant to invest and implement such solutions as they face the dual challenges of differing customer requirements in terms of technology and process as well as the lack of a coherent proposition that reduces airline costs while at the same time maximises value for the airport. Global industry standards help to minimise the impact and are at the core of the Fast Travel solutions offered by the service providers and vendors.
IATA Fast travel strategy

Part of the problem is the aping of consumer technology, but always being behind, so user experience is not consistent with the expectations set. Consumer replacement cycles are within two years but enterprise IT roll-out projects still take that long to complete; so it stands to reason that they will always be behind the curve. 

Two trends that will help here are firstly modular systems using off the shelf components pushing the "specialised" part of the system into software. The second is BYOD, having dumb infrastructure that again pushes the business rules and experience into a place where it can be easily updated to keep pace with consumer expectations. I don't think iBeacons and apps are the answer just yet, but I do think that they show a glimpse of the future way of working. 

Airlines probably stand to gain the most here from reduced ground handing cost and the airports that are going to be the laggards are in poorer areas. So it seems bonkers to me that the question of "who owns the customer" is heard as a blocker to both parties working closer together. 

Isn't the better question "what provides the customer value?" - so do more of this - and "what detracts from that value?" - so do less of this if possible. I saw a great example of a creative answer to the second question at Disney World, where quite young children can be made to queue for an hour during busy periods. So, they have themed play areas with buzzers to maintain queue position in the busy rides. So the children are kept entertained and out of the sun. 

I'll sign off with a couple of questions ... What would a creative solution to this look like for say security queues? How could airlines share more data with airports to benefit from reduced costs?

Tuesday, 25 April 2017

What I've been reading w/c 17/04/2017: Diversity and Ethics

Slightly delayed post due to going on holiday!

My week's reading started with Hofstadter’s Law and Realistic Planning By Jane Collingwood where she outlines how "pessimistic-scenario generation is not an effective de-biasing technique for personal predictions.”  this got me thinking about how much of what we do in our lives is shaped by the people who provide us with the services and products that we use.

This article by Monzo is a perfect example of ethical product design should be done. They have thought about what their mission is, who their users are and what issues they might face. Compare and contrast with stories of how Uber use psychology to exploit drivers to see the negative face of "disrupting industries" when that is the sole aim. 

Again working conditions can have an impact even in subtle ways. There is a case here for Product Design and OS professionals to  provide more support on reporting usage to users. Computers are much better at this then humans! Could more thinking like this be a baby step on the path to more development teams thinking like Monzo? 

Next on my reading list was a great post on the difference between basing your business around the vocal 5% and really using feedback for process improvements. So, it's important to make sure to include diversity in customer feedback as well as on your team to avoid systemic bias.

This systemic issues is why diversity in IT is so important. As algorithms control more of our lives the bias of people writing them matters -  the world economic forum has also come up with a list of 10 ethical issues with AI  one of which can be summed up by Blay Whitby "Autonomous Vehicles don’t need to solve the Trolley Problem: but maybe we do" ... which means for a fair outcome for all of us, we need to make sure the people solving the ethical issues are representative of who they will impact. 

Which led me to a good write up of recruitment and diversity by @ashedryden: The Ethics of Unpaid Labor and the OSS Community -  As open source contributions are used as part of hiring process, we need to be careful this doesn't lead to an echo chamber of sorts. Ashe explains this much better than I could so well worth reading her work. 

Looking around at the different groups product design needs to consider I was pleasantly surprised that Microsoft has an awesome body of knowledge on inclusive design. 

Finally Important to remember Women’s History as part of Silicon Valley. There is more diversity than we commonly hear of. The current status quo isn't a foregone conclusion, so it's in our power to change it! 

Thursday, 13 April 2017

"Untitled side project"

So, I've mentioned my side project a couple of times I should start to elaborate a bit more. I won't share the name or details just yet. Not because I'm worried about someone stealing it. Rather the idea, hypothesis, and validation are the next installments of this story. When embarking on this journey as a side project it had to be fun and interesting. A chance to try a different experiment to my day job. Different tech stack, different tools. Also be of use to people. So I set to work with a collaborator on something to meet these needs and we discussed a basic idea. The first thing was creating a Trello board. One of the first cards was "think of a name". Next came looking at the front-end stack that was lightweight and easy. A PaaS provider that had a free option and integration with source control. This also needed a free private repo option. And it was vital to have a pipeline that built and deployed a working version of the code on commit. Luckily GitLab and Heroku play nicely together. It seems a little too easy! As an added bonus the Source app means a code change and deploy to a working server can be done on my phone - seems ridiculous but that is easier than updating this blog! One of the key early decisions was "NO PASSWORDS". There are more than enough sites that provide identity services, from GitHub to Google+. The last thing anyone needs is another username and password to remember.

Then in a flash of inspiration, the "think of a name" card was completed! A domain name registered! A Twitter account started! A user journey mapped on the back of an envelope! These were exciting times in "side project land".

Then ... nothing much. Life got in the way. Holidays, illness, and the day job. I would imagine this is a key struggle for many startups and side projects. Maintaining momentum. We found ours in targeting a landing page to launch a user research survey. Then booking in a night to work on it, to make sure it happened. With an actual site and the prospect of visitors. Another no brainer was the importance of analytics. With so many SaaS providers that have starter plans for free there is no excuse. So far we have and Google Analytics. Pirate metrics here we come!

Sunday, 2 April 2017

What I've been reading w/c 27/03/2017: Chatbots and AI

Chatbots are an interesting example of how supporting technology can be the catalyst for innovation. In this case smartphones with messaging apps, constant fast network connections, and an API economy. All these enable comparatively low-tech chatbots to be viable (even I wrote a production text interface to an asset DB in 2000!). So, it's not new technology that's the innovation, it's combining existing technology in ways that change behaviour.

This article in the Harvard Business Review is spot on. AI systems much more like employees than traditional IT. Because it "learns" you often can't just lift the data/business rules into a new system. So you need to start thinking about "handover periods" and "training" a lot more. (Also make the jobs-to-be-done framework about hiring tools to do a job much more apt!)

There is much promise in Artificial Intelligence, this article on what to think about machines that think contains some views around that which don't seem to be common outside academia or TED talks. (So possibly useful for those looking at AI without that academic connection, definitely much more informative than a lot of mainstream press coverage)

Time to go beyond mobile first to AI first products, different personalities needed for different user experience (UX) that the AI driven product will provide. This needs to be considered up front, in the same way that moving from desktop to mobile needed a mobile first approach. 

Interesting that from a marketing POV there is a lot of AI technology around but not many integrations in use. Maybe they need some tools like Wrappup?

Over on Medium a brilliant example of going from technological differentiator to commodity item. Reminds me of a friend's complaint years ago that his hard-won Flash skills were now standard tools. Same thing now with AI, what I learned to hand code is now a cloud-based resource. At least I will have appreciation of how it works (and pitfalls in training ;) 

Finally, Fred Hsu CEO of explains how for SaaS and other tools AI is not only going to change the way we work but also the way our tools and services are priced ... there could be opportunities with both. 

Saturday, 25 March 2017

Tools for taking notes and sharing on the go

Moleskin Smart Writing Set 
As a product person, I am always thinking about unmet needs, ways of working, and pain points. One of those in my professional life is taking notes. I have tried Evernote on a tablet, into Word on a laptop but always come back to pencil and moleskin notepad.

This then is a bit of faff in finding notes at a later date or sharing with colleagues. Scanner apps that create PDFs help, but I usually forget and take a photo that I then ping off by email.

Imagine my excitement when I heard about the reMarkable tablet in development:

The paper tablet for people who prefer paper. Here to replace your notebooks, sketchbooks and printouts. Paper-like reading, writing and sketching with digital powers.
Sounds great! This has the potential to do for note taking what the Kindle did for eBooks. But it is still only on pre-order and my birthday was this month. So for around the same amount of money I have got two tools.

The Moleskin Smartpen with "paper tablet". This is a smartpen with a moleskin pad that has a tiny pattern printed on each page. The camera in the pen picks up where the strokes land on the page to store in memory. This is then uploaded to an iOS or Android app via Bluetooth.

I have only used this for a short while but what I like already is the blend of analog and digital. It doesn't need a change in working behaviour for me to get some benefit. It also degrades gracefully. If the pen runs out of ink then you still get an electronic copy. Also if the battery dies, still have the paper version.

One neat feature, if you have the app open and place the pen on the mail icon in the top right corner. It will then export the current page to PDF and attach to an email. Not a massive time saving, but a nice connection between the physical and digital actions.

The people behind the smartpen technology - Neo - also have a section on their site where you can download NCode PDFs. This is the format that the Moleskin pads are printed in. This allows you to use full A4 pages with grid, line, and graph patterns.

Another thing that I hadn't considered before I uncovered during setting up the auto sync. With Evernote, Google Drive and OneDrive all supported I get redundant backups for free. This also avoids vendor lock-in. Both things that enterprise internal IT departments tend to love.

Acer Chromebook 14
Following on nicely from this last point is the second new tool I am getting to grips with is a new Chromebook. This is a cloud-native notebook, with limited disk space and based around the Google Chrome browser. Like Microsoft I am a bit late to the cloud party, having grown up with Unix and DOS command lines I'm not afraid of file systems. What struck me was how easy it was to setup. It even brought in my bookmarks from my Android tablet. 

In fact, it was surprising how much was already setup from details used on another device type. For example, my LinkedIn account details were saved in my tablet, so didn't need to log in when accessing site for the first time.

The specs would probably be quite low for a Windows but I find it speedy. One annoyance is a general one from the Google/Android ecosystem. Finding apps that "support" your device. This isn't really an issue in the iOS or OS X worlds, and one of the reasons I stayed with iPhone and Mac Mini. It would have been nice to use the Moleskin M+ Notes app on my Chromebook, given its long battery life. I am hoping that they add Android app support to my Chromebook soon. I don't suppose I need to worry too much with the range of auto sync options available.

My next task is to investigate tools like:

As I see how much of my side project can be done not only using free tools but also a largely online and cloud-centric approach.

Sunday, 12 March 2017

Welcome, a re-introduction to my blog

I started this blog over 5 years ago now and it feels like an exciting new chapter in my life is beginning, so thought I share some of this and what it might mean for my humble blog. I started to write about my work and how what I saw measured against the theory.

Although work is an exciting place at the moment. With three projects due to go live in the next three months that will complete the vision that has guided me in the past 2 years. I won't write about them too much here as we now have a much more active work blog that I'm scheduled to write on. Some of the things that we have discovered during this work will lead to new features that I'd also love to talk about, but need to keep under wraps until we are ready to reveal them!

So what will I be writing about? Well, more book reviews and more meetup reports. Also something personally exciting as I am starting a side project that allows me to apply lessons learned from HBX CORe, Impact Mapping, Jolt by Justin Jackson and storytelling in a different vertical from my day job. It's very much a B2C project rather than B2B product, which is also fairly new for me so I might write about any differences that I find.

One major advantage is that I'll be able to write about much more specific examples and lessons learned ... plus anyone who is interested will be able to go and use the product :-) So, I hope that you will continue with me on this new journey.

Further reading

Wednesday, 22 February 2017

BOOK REVIEW: Fifty quick ideas to improve your user stories by Gojko Adzic and David Evans

BOOK REVIEW: Fifty quick ideas to improve your user stories by Gojko Adzic and David Evans

It has taken a while since my last book review to get the time to do much reading with moving. But here is my round up of "Fifty quick ideas to improve your user stories" by Gojko Adzic and David Evans

Formats: Paperback, ePub, Mobi, PDF

Where can I get it? From Leanpub, Amazon or any good bookshop.
Who is it for? Anyone involved in a software development project working in an iterative manner. As long as they understand some of the basics around user stories, e.g. they know what INVEST stands for.
What's it about? As the title suggests "how to improve user stories", but it is a bit more than that. It covers the whole process including planning and iterative delivery activities.

What's the book like? Each double page spread follows a similar style starting with an introduction to a new tip. Often illustrated with an anecdote from the author's experience. Next is a description of the key benefits of the idea behind the tip. Finally, it finishes with some practical ideas on how to make it work. This takes the theory and presents it in ways that you can apply it for whatever you are working on

The chapters follow the life cycle of a user story:
  • Creating stories
  • Planning with stories
  • Discussing stories
  • Splitting stories
  • Managing Iterative Delivery
The story splitting chapter was my favourite section. These tips cover some of the real gnarly issues in non-trivial agile projects. For example, putting off implementing a reporting system until quarterly report is due and developing the key functionality generating the data as text files. These can then be imported to a new infrastructure once ready.

The tip introductions are really good at giving you a grounding in the context that Gojko and David have developed their experience. This is an important factor as it allows the reader to gauge what factors are similar in their context. All forms of advice (otherwise known as "best practice") need to be adapted to the real world context they will be used in.
To sum up, lots of practical tips to help get value delivered. For example, don't get stuck in a rut with stories that aren't appropriate like technical tasks. If you want to know more about the book topics then check out this cool mind map of the book.

Sunday, 12 February 2017

MEETUP: "Collaborative product management " at ProductTank Brighton

Looking at my notes I forgot to blog about November's meetup "From Startup to Corp: The differences, and how to adapt to the change." sponsored by 15below. There were three very good talks on the difference between very small companies and massive ones. Some highlights from the three talks:
  • Don't feel like you have to aim for moonshots if you aren't Google Ventures - you only get a limited number of bets compared to GV, so a 10% incremental improvement every year is nothing to be ashamed of!
  • Don't feel too embarrassed by your peers - if you start of at the same time as JustEat but have a much more niche market it's OK to have more modest growth.
  • Just because you can see everyone doesn't mean that they know what you are doing - large organisations have the privilege of knowing communication is poor so actively do things to encourage it, in a small startup you have to make ensure people are informed.
  • Large organisations can offer more support and flex around you - for example cover maternity leave or secondment to another business unit.
But anyway ... last week was another interesting edition of the ProductTank Brighton meetup hosted by Rakuten on "Collaborative product management".

Unfortunately, the first speaker scheduled - Dan Kidd - couldn't make it to deliver his talk "Now That's What I Call Continuous Delivery". I was really interested to hear how madgex have used the improved feedback cycles. I am looking forward to the reschedule!

Instead, the first talk was by Ed Vinicombe on "My three critical values of collaborative design" and how traditional ways of working are changing in knowledge work. He is really lucky at to be guided almost entirely by user research (and not HiPPO!). It sounded like this gave him the freedom to collaborate around the requests and deliver quicker.

Next up was Mike Rowlands talking about the development culture at LShift. Maybe controversially for a Product Managers meetup he was suggesting that the role might not be needed! They had very deliberately set out to recruit senior polyglot developers. Ensuring that they fit working as lead developers in an AgilePM environment. Both the team sizes and projects undertaken are kept small. It is up to the lead dev to choose the appropriate tool for the job and peer review is used as a check and balance. They work closely with the clients and do a lot of the communications without an intermediary. This collaboration between client and dev is what helps develop the product. I can see how in this specific environment it works. As it is appropriate in consultancy style "products", but I'm not sure it works as well for long running commercial off the shelf software.

It was interesting hearing from an MD who had created the kind of company he wanted to work in and vigorously defended that culture as it expanded. It sounded like being careful at the type of work accepted played a key role in this. A final little tip was that they aimed to get the work completed and handed over so the clients could support their own BaU operations. But the clients in most cases kept them on support contracts because of the way they solved problems in a transparent manner.

Monday, 6 February 2017

On data blind and data informed

This post is the story of a personal journey. It starts with a legacy product. Both powerful and flexible having grown to meet needs over time. But it is my view on steps to improve the process in my context - so your mileage may vary.

Phase 1: Transactional DB as source

Started with looking at the transactional data. Some things were obvious, one option in the UI is recorded as true/false in the DB. Some were more complicated and required more investigation. To back this up we conducted a user survey. Followed up by a workshop to find the most important user journey to concentrate on. This highlighted three related journeys. The key characteristic of activity in this phase was mining different structures of data. The format of the data coming from custom reports.

Getting to the questions to ask was guided by a user survey to almost the whole user base, a smaller workshop looking at the "job to be done" for the platform. Other than that the skills needed where accessible to most development companies or teams - SQL and Excel.

Phase 2: Audit metrics

During developing these we put in place metrics to track the different activities. This was across all clients in on pace. Even though these were basic metrics they did allow patterns of usage to emerge. For example, actions being triggered and then canceled. Repeatedly. We were able to follow up and check why this happened. Before it became a support or account management issue. This phase had standardised event data that was easily aggregated and sorted. But like the previous phase still tabular in nature. A key benefit here over the transactional data was explicit time stamping of events and users.

In this phase, we also started to use Google Analytics to make the collection of browser user statistics easier. One advantage that we gained over web server log file analysis (other than having to do log file analysis ;) was the browser window size and native screen size. This was useful for guiding the constraints on page design, and tracking how this changed over time.

Phase 3: Users behaviour

The next phase came in adding in more user journeys and variations on the three that already existed. the different combinations made it more difficult to see what was happening and who the interrelated functions were used in tabular data. So the next step was to use session recording. This also helps speed up doing usability testing by opening up a way of recording sessions and playing back. This is invaluable as the pace of development in the new framework speeds up. The feedback on how that solves the problem also needs to scale and improve in quality.

This does bring interesting challenges on workflow. Previous phases have ultimately ended up in some for of document - Excel spreadsheet or table in Word. So a factor in evaluating tools is "how can I make sure this interesting behaviour can be referenced". Can I get a permalink to go to a point in the playback? Will it integrate with our chat tool to let people know something has been found? ... will they then use it?

Phase 4+: Infinity and beyond...

What comes next very much depends on which process problem we are looking to solve next. One thing in our context that probably won't make sense is A/B testing, as in our B2B environment consistency and predictability are valued. Some of the things that it might include are:
  • ROI - Taking into account factors like whole-life costing of the new product development, revenue generation/protection and feature usage to get the most bangs per buck.
  • Process mining - looking at user behaviour to find correlations between tasks to form new packages or products.
  • In app feedback - fine grained/easy feedback, like the MSDN "was this article useful" to flag issues.
  • Visulisation - how information flows through the system, maybe animated to really tell a story.
  • Heat maps - Another way to tell a story about feature usage.
  • Further development of customer types and tailoring products for them.

Further reading

Monday, 30 January 2017

On Nike+ and rebirth

The Nike+ running app underwent a drastic transformation to become Nike Run club. Gone is the old red and white branding, replaced by black and volt yellow. Impeccable timing as Brighton and Hove Albion switched their away strip to the same colours (with Nike as the kit supplier). The whole focus of the app appears to have altered from a useful run tracker in a "quantified self" category. To a more limited - albeit focussed - coaching app.

This wasn't the most popular update if my friend leaderboard is anything to go by. Shrinking from 6 regular people every month to me and one other and even then only on some months. I wasn't running regulatory when the change happened but remember some complaints about users losing their training plans. We runners are a funny lot. To illustrate, I had a conversation at the weekend with someone who was upset about running a marathon in 4 hours and 2 minutes rather than just under 4 hours. Our training plans are the rock we build the performance to get our desired times upon. So bad move!

Example of old website
Things I have missed/found harder to find:
  • Achievements
  • Comparison of your effort in past month vs cohort or general pop (on the website)
  • Stats on split of runs by time of day, day of week etc (on the website)
  • Use of average of previous 7 runs in predicting time or distance when setting up a run
  • Use of music controls to pause/resume a run
  • Community .... no one is there anymore!

Example of achievements

Looking around the website I found that it is still possible to see the breakdowns I mentioned above. But they are hard to find now, where before they had been on the first screen. The average of the last seven runs has now moved from the app to the website. The same with the achievements. They used to pop up at the end of the run like the, now you have to go to your settings and look in the achievements menu.

Example of new sticker
Things I do like...
  • My coach program
  • Stickers :-)
But there are plenty of coaching apps already. Ones that don't deprive users without access to their training plans! Albeit my liking Nike Run Club isn't based on previous experience with app based coaching. I like that it is a personalised plan, with targets that adapt as you train. I have already had the plan update once; stretching the pace of speed run sessions. This is helpful in feeling progress and motivation to run on cold dark nights. Finding the coach is also now a larger part of the experience and the quick start for runs makes my life easier. Again also good for cold dark nights when I don't fancy running!

The stickers are amusing for a bit. Finding the related stickers that only appear on Sunday made me chuckle, but it isn't tapping into the stats and comparison side of "social" running. I'm not sure if it is me, but it doesn't appear to be as easy to share the link the to run. Peaking at the route and splits of other runner's workouts is key aspect of this comparison.

There is one area that I don't understand. This isn't their core business. Yet they have missed extending support for the shoes and running gear sales. This would tie the app into their core business a lot better. Plus it would differentiate Nike+ from the other coaching apps. They collect lots of information about people, how often they run, where they run, what the weather is like, and which shoes they already use. They could have a bit more of a soft sell. So rather than reminding you that your shoes need replacing. They should build on this to suggest some alternatives and deals. This is the opportunity to even cross-sell, for example winter running gear for trail runners. 

But hey, what do I know. I don't know what their product strategy is. Or what their OKRs might be. Just from the outside, as a user, some of the decisions seem puzzling.