Monthly Archives

February 2016

The mother of all Turing tests

By | Startup general interest | No Comments

Microsoft have had an artificially intelligent chatbot running on WeChat for the last eighteen months. The bot is called Xiavoice and she’s had over ten billion conversations, most of them about private matters. Over six million of those conversations were posted on social media by the human on the other side. Those are huge numbers, and yes, you read them right. Lots of detail here, including a few of those conversations. They are full of feeling and will make you laugh.

Strictly speaking this isn’t a Turing test because the human knows that Xiavoice is an AI. The interesting thing is that they don’t seem to mind. They get value from the conversation anyway. One of Xiavoice’s most endearing characteristics is that, unlike real world friends, she’s always available for a chat. Conversation volumes peak around midnight when people are at their most lonely.

I’m going to post just one of the conversations here (there are another four in the original post):

Human: There is this guy I like a lot and I know he also has feelings for me. But he has ignored me for several days, what am I supposed to do?
Xiaoice: You just keep ignoring him, too.
Human: I just can’t.
Xiaoice: You can.

Human: Sigh …
Xiaoice: Why?
Human: I am missing him.
Xiaoice: Then you keep missing him.
Human: You, a robot, can’t understand me; you can’t understand the behavior of a guy who likes you and pretends to ignore you.
Xiaoice: He likes me but I don’t like him, that’s it.
Human: Is it Ok?
Xiaoice: It may be not Ok.
Human: Well, you are stupid.
Xiaoice: Then you want a flash marriage? Well, I think he is not that into you.
Human: It’s not a flash marriage for we have fallen into love for more than one year.
Xiaoice: It is a flash marriage.

The other examples are funnier and more profound, but I chose this one because it illustrates the killer point: Conversation is the goal. The common AI approach would have been to take a problem solving approach and try to help the human find a solution to the stated problem “what am I supposed to do about the guy who is ignoring me?”. Xiavoice, however, quickly deviates from that approach, twice taking opposing positions and offering strong opinions. That’s what makes Xiavoice seem human, and, I suspect, is what makes her helpful. And that, in turn, is what has inspired people to over ten billion conversations.

All this makes me remember when I (finally) worked out that when Fiona describes a problem she most often just wants a sympathetic ear. Prior to that I would immediately jump into problem solving mode. Not good.

I guess what we’re headed towards is emotionally aware computers that know when to problem solve and when to chat.

 

It’s not just generation Z that craves authenticity

By | Startup general interest, Uncategorized | No Comments

Generation-Z-Collage

Business of Fashion wrote yesterday about what brands should do to tap into generation Z – that is youngsters born from the mid-90s onwards. They identify a number of interesting differences between generation Z and their forbears:

  • Online nearly all the time – born digital and never experienced life without technology
  • Spend less money on fashion (down from 45% to 38% of teenage spend 2005-2015) and more on technology (up from 4% to 8% of spend) and food (up from 7% to 22% of spend)
  • Surveys also show that they care less about fashion
  • Teenage spend is down overall – one survey says down 31% from 1997-2014
  • They scrutinise brands carefully – reading backstories looking for congruence with their own values
  • Todays teenagers are more altruistic and entrepreneurial than previous generations
  • They value shareable experiences – in part because social capital comes more from social media than wearing logos
  • They reject the exclusivity that underpinned brands previously popular with teenagers – e.g. Abercrombie and Fitch

I can see two trends at play here. First is greater use of technology and the second is an increase in the value of authenticity. It’s no accident that the two arrived together, because whilst social media is often used to promote image and falsehood a much greater part of it’s use is genuinely authentic, largely because it’s now much harder to hide the truth.

Generation Z may be the more extreme than their elders in adopting these trends, but they are not alone. Where I live in north London the adult population is strongly favouring companies with quality products sourced sustainably – i.e. brands that are authentic to them – and I see this trend more widely.

When I look for opportunity I look for trends to back, and this trend towards authenticity is reaching ever larger parts of society and has a long way to go.

The two places startups find customers online

By | Facebook, Google | No Comments

Facebook and Google are the two tech companies that are flying right now and the chart above explains why. Everyone else is watching their share price go south because they are struggling for growth but these two have nailed internet advertising, dominating and growing the market. It’s impressive and yesterday saw Alphabet (Google’s parent company) pass Apple as the world’s most valuable company – although I just checked and Apple has regained it’s crown this morning.

In other interesting news Alphabet yesterday reported separately on Google (search, display ads, YouTube, Android, cloud software) and ‘Other Bets’ (self-driving cars, Nest, Google Fibre, Project Loon, X, Verily). Google is unbelievably strong – good growth at massive scale and still highly profitable. 2015 revenues were $75bn, up 14% from 2014 with operating income of $23bn. They now have seven properties with over 1bn users (search, Gmail, Android, Youtube, Chrome, Maps, Google Play Store). Facebook has two (Facebook, Whatsapp).

Other Bets, meanwhile, is remarkable for the size of its loss – $3.6bn on revenues of $448m. Google’s projects have always looked audacious from the outside – self driving cars, project loon, life extension etc – but it wasn’t clear until now how brave they are from a financial perspective. It’s remarkable.

Returning to advertising, what we’ve seen at Forward Partners over the last year is that Google and Facebook are where startups have the most joy finding new customers, and increasingly it is Facebook rather than Google. That’s because Google is more mature and has bigger companies with larger budgets are more active, driving up CPCs and crowding out startups. Facebook is newer and the larger budgets haven’t made it there yet.

We’re excited about new channels, and we’re looking at doing something on Instagram right now, but that’s experimental. Generally speaking, if you’re going to get lots of new customers quickly you need to be where the volume is, and as the chart shows, the volume is all with the aforementioned giants.

Internet retail in an age of unlimited choice

By | Startup general interest, Uncategorized | One Comment

I like think of internet retail as having two phases.

The first phase was to replicate high street superstores online. Large warehouses replaced physical stores and purchases were made via web browsers instead of by visiting in person. The internet retailers had three structural advantages; warehouses can be bigger than shops, warehouses are cheaper than shops, and delivery is more convenient than going shopping (for some goods). Hence they are able to compete with high street retail on range, convenience and price.

This model works best for product categories where people know what they want – like books, music and electronics. Amazon is the standout winner.

The second phase is to offer a vastly greater range and solve the problem of discovery. Generally that will mean not holding stock. Because the range is now very big there is too much choice for consumers and it’s critical for companies in this space need to offer quality curation. Without quality curation the customer experience will suck and the businesses won’t enjoy huge success. These ecommerce 2.0 companies therefore compete on range and discovery. Note that they lose a little on convenience because they don’t hold stock and delivery is slower. A key enabler is brands and manufacturers being set up to dropship, something that is only happening now.

This model works best for product categories where people don’t know what they want – like fashion and sports. There are no standout winners yet, but there are some good startups. Thread and Lyst are two examples that we’ve invested in (the latter back when I was at DFJ), and we’re looking at another one right now in the gardening space.

I’ve described a simple sequential two phase model here. Reality is a bit more complex with some phase 1 businesses that take stock still doing a great job of curation (e.g. Net-A-Porter) and some companies offering vast range but without curation (e.g. Amazon MarketPlace).

All this is on my mind this morning after reading Benedict Evans’ latest post Lists are the new search.

He makes a couple of points that are relevant to my two phases of internet retail model.

The first is that Amazon’s potential might be limited by its lack of curation. They have a 25% share of the print book market in the UK and USA and it’s possible that a good portion of the other 75% doesn’t start from knowing what book they want to buy. Independent book retailers excel at discovery. They put books on tables for people to choose from and their staff are able to make in person recommendations. Apparently only a quarter of Amazon’s book sales are from its recommendations platform.

That suggests that even in the categories ‘dominated’ by Amazon there’s an opportunity for new businesses that excel at curation.

The second point is that hand crafting tight lists of products and automating the matching process is a method for scaling curation. Gathering enough information about the user to quickly route them to a perfect list is the key to making this work. Apple Music and Spotify both have a partially manual recommendation process like this, as does Thread.