Monday, 30 May 2016

CONFERENCE: The Lean Event, Brighton and Phocuswright Europe, Dublin 2016

I have been struggling to write a summary of The Lean Event and Phocuswright Europe as they both packed in so much content, I have so many notes to read through! Taking the two together it's clear to me that they are natural complements. Indeed Umesh Pandya's talk on "Learing to build wayfindr: independant travel for blind people" would not have been out of place at Phocuswright, just as Gary Morrison's afternoon keynote on Expedia Worldwide could have been a Lean Event session on lean in the enterprise. So i'll pick a couple of sessions from each to talk about.

From The Lean Event 

There were so many good sessions over the two days, but I'll briefly talk about Jared Spool's keynote on "Building a winning UX strategy" for this insight on Innovation alone - innovation is the space between current experience and aspirational experience. The simplicity of looking at innovation as gap between frustration and aspiration/delight was quite surprising.

Jared also put forward the idea that roadmaps should focus on when we are going tackle customer problems not the technological solutions. He showed in using a journey map to show areas of frustration you can then target these areas and get them into the delightful zone.

I also chose this session as Jared also had a few examples from travel and airlines in America, so was of particular interest to me. There are so many systems and processes involved in taking a simple journey, especially one that has been disrupted, that the service design needs to consider the user expereince as a whole.

As an added bonus The Lean Event had a table where you could write yourself some takeaway comments as a reminder. They then posted these 3 weeks later, as it turns out that was a good time after the conference to be reminded! Recent enough that you can remember what it means, but long enough back in your day job that you might need a nudge.

Phocuswright Europe

I can't remember who said them but my two favourite quotes from panel sessions at Phocuswright were "is this a startup or a feature?" and "OTAs are tech companies" (Similarily "Ryanair is a platform" and "JetBlue is a customer service company that just happens to fly planes."). There is a strong relationship between these quotes although they cam efrom different panels. If your startup is just a feature OTAs should have, then they will replicate it if they can as this is their core business. I have picked two pitched to try and  illustrate this.

Elevator pitch - Packaging golf tours dynamically - this solves the problem of time constraints and hassle for consumer and as a distribution platform for businesses. Using ther widget means everyone can become a golf tour operator - without human intervention or industry knowledge, so can concentrate on their core business. 

This feels like it fails the feature vs startup test. The core business of OTAs is literally a superset of this feature, with this being just one "personalisation". The OTAs will have an advantage in the long run with greater reach, development resources and benefits of scale. Remember "OTAs are tech companies".

In a nutshell, the product scans through e-commerce sites and finds broken booking journeys to see where the problems with conversions could lurk. It scans pages to collect analytics info, rather than rely on tags in the link. Then compares which tags are there/missing/misconfigured. This runs as a SaaS platform to do the testing with no code to run on site. 

What I liked about this pitch was that the CEO has brought her experience from being a marketing manager, and the pain of the kind of reports that she had available, then she has created a product to fill the gap. My opinion is that this is not quite the core business of the current providers and a complimentary product, so more than just a feature.  this is even a product and service that a tracking provider or a large industry client like an OTA could use.

EDIT: Talking of features dressed up as products, after Phocuswright I became aware of an EU funded research project called EuTravel.This aims to a create multi-modal travel planning tool that seems just like Rome2Rio. After a further look at the project site I can see some science in there... however, mostly it appears to be subsidising competitors feature development. They could have worked with existing providers to collect data on usage and social change, if that's what they need, and I would've thought that this might be more cost effective. (As a bonus they could also have seen how different approaches of these partners changed things)

Further Reading

Friday, 27 May 2016

On AI and hype

Machine Learning Miller by
When I wrote about AI and the Future last year I was reasonably excited, as an Artificial Intelligence (AI) graduate, of the possibilities and jealous of those beginning their careers in AI. Since now they have the luxury of extreme computer power and storage, 3D printing and the other abundant pieces of technology needed to create the future bounded only by our imaginations!

The past couple of months though and I am noticing a bit of a trend in conference presentations (and tweets coming out of conferences) that seem to have moved a lot of the hope and hype around big data onto AI. Or more specifically machine learning. I am not going to single out any specific examples, but I feel this covers two basic areas:
  1. I don't need to know about my data or structure it to get useful information and 
  2. I won't need to configure things. because machine learning.

(Lack of) Data structure

I am not sure what is driving this need but I guess it has something to do with the falling price of storage and the rise of world wide data collection via the web. So lots of business now have lots of data sitting there. There is also hype about the businesses that do data analytics well - e.g. Amazon or Uber - that leads people to think "They make money from their data. I have data. I should be able to make money out of it".There are also lots of blog posts and talks about this tool or that (e.g. Hadoop or Map Reduce) has enabled the data analytics, which usually gloss over the basics like linear regression because everyone does that...right? 

In the Computerworld article Machine learning: Demystifying linear regression and feature selection they make this very good point
Much of the art in data science is understanding the problem domain well enough to build up a clean set of features that are likely related to what you want to model.
 and show how using training data that has been modelled with clean features outperforms using all the data without any domain insight. So yes, machine learning probably can be useful just be prepared to understand the data and be able to run your own regression analysis on any training sets needed.

Configuring things

Again, the hype seems to have come from wanting to copy successful market leaders and I can almost hear people thinking "Netflix adapt the recommendations without configuring it, can I configure my users menu structure? (or whatever)". Adapting to changing usage or conditions that your software operates in is a perfectly reasonable desire. But here it's worth asking how much "adaptability" to we really require? and how much of the conditions of this can I realistically foresee? For example adaptive polling of mailboxes may not require a vast neural net to see traffic cycles and usage behaviour to tweak the likely poll time. We might just be able to write much less code that has enough adaptability based on what it found last time and a simple heuristic.

Final Thought

So please, do explore the potential that machine learning has to offer but don't expect it to be a silver bullet or replace understanding your domain. I think that AI surrounds us to much now for another AI winter to set in, but too much hype isn't good for any tool or approach.

Monday, 9 May 2016

On Twitter and lists

Being into product development I am unsurprisingly fasinated by features and how other product people approach and solve problems. One particular feature that intrigues me is Twitter lists.  It is a simple feature with one configuration option - private or public - and one setting to either add people or remove them from the list. There are no big built in workflows or obvious big assumptions (at least to me) of how Twitter are forcing intending you to use their list. One of the things that I love about this is that it allows you to project your needs onto the feature and use it to solve your problem.

So being interested in finding out more about this - and wanting to test out Twitter polls! - I did a quick survey to see who else uses lists

Now it's not very scientific and the results were hardly overhwelming, but it did reflect a straw poll that I did in real life. Not much interest, but if someone was interested then they not only used their own lists but subscribed to other people's as well. Looking around I saw some interesting policies to get engagement based on following behaviour, e.g. This blog by Edward Nevraumont, so I wondered how lists could be used in a similar way. I have slowly developed a usage of lists to help me at conferences and it breaks down into three phases:

  1. Phase 1 - pre-conference

    As the conference organisers start to tweet including speakers twitter handles I create a public list for that event and add any accounts to it (including the organisers).
  2. Phase 2 - during the conference

    This is especially useful for multi-day congferences where I follow the conference hashtag for at least the morning session of day 1. Adding new accounts that I see pop up either on the hashtag or retweeted with the hashtag. This means that I am ready for day 2 to see what the other attendees or speakers are discussing and catch comments that may not have the #hashtag included (or a slight typo in the #hastag! ;-)
  3. Phase 3 - post-conference

    I now have a tailored list of people who are pre-qualified as interested enough in a topic to not only attend a conference but tweet interesting snippets on that topic. Even if I follow the people on this list separately it can be useful to quickly dip into likely conversation areas.
I have tried this a couple of times now with some conferences easier than others, for example The Lean Event (Brighton 2016) had mainly individuals attending who were active on Twitter is different from a trade show like Phocuswright Europe (10-12 May 2016) where my list is taking shape and which has more corporate accounts tweeting.

My lists are usually fairly plainly named in terms of intentions, just the event name. But I have seen some which are more of an invitation, for example I was recently added to this list inviting me to a face-to-face meeting at MTPCon ... unfortunately I wasn't there and only retweeted someone who was. This is where you need to be careful. Usually in my usage that's OK as we are interested in topics of interest rather than physical presence.

That's how one way that you can use lists, how do you use them?