Characters as a Service: Paying for personality in our products?—?Part 1?—?Bots

Characters as a Service: Paying for personality in our products?—?Part 1?—?Bots

 

Characters as a service is a complicated idea I am coming to define. I am writing about it before it is fully formed so that I can explore: the premise, examples, case studies, and make proposals for the future. I hope you enjoy.

We’ve seen a lot of bots launch recently, with likely more on the way with the announcement from Facebook, and the thriving platform pioneered in Asia with WeChat. It’s actually kind of exciting. I think a about interaction, and experience a lot in my line of work, and this seems like a smart next step in the evolutionary process of an older, still fantastic idea.

The premise with most SMS bots is simple. You text a phone number with a question or goal in mind, and someone texts back ready to help. Only it’s not usually a person, it’s a bot that took your question and routed it through a series of options until it got to an answer that made the most sense, and it sent a pre-programmed response.

With so many people writing about bots one or two is bound to call them AI, after all, they are bots, and they are talking to us right? It’s a bit of a misnomer to call it AI, but people do anyway. I’ll admit it makes it sound better, but the idea is pretty cool as is even without the embellishment. Sure they are bots, and they are talking to us, but it’s totally cheating. Bots are pretty much just productized if-then statements. So then, what can you do with these bots?

AI is coming. While bots are simple right now, AI is on the way, simple AI of course, but AI none the less. AI that has been trained with millions of interactions and counting. AI that becomes if-then statements that can discern. If-then statements with context, and memory. If-then statements that approach something close to intuition with regards to certain tasks in certain narrow bands of expertise.

Regular vanilla if-then statements can do quite a bit now though in the meantime. You can ask for a particular pair of shoes, and it can find them for you. Need directions, an uber pickup, a sandwich? No problem, there’s a bot for that.

Need something? No problem, there’s a bot for that.

But bots right now can only do so much. Which is why, when you text Operator, you get a person. Even thought it’s a person, they operate much in the same way as a bot. They perform a task they have been ‘pre-programmed’ to do, only they can be a little more friendly about it. They can go ‘off-script’, to make every time you use it, new and interesting, instead of strict and predictable.

If the nature of the job to be done is very specific, with more exceptions than rules, it becomes very costly to build out the logic. Bots are most cost effective when very similar tasks need to be performed regularly. The responsibility of the bot, in that case, moves from problem solving, to more a matter of improving efficiency.

Sometimes it makes sense to work with a person, and still other times favor working with a bot, but I actually think there is a sweet spot in between that companies will start to explore very soon.

I think there is a place in between ‘hard coded’, and ‘hot blooded’, a place between silicon and sinew. I’m talking about a programmable personality. Kinda like AI, but very niche AI. These personalities, or characters I’ll call them, can usher in an era of appless interaction that people truly love to use.

I think there is a place in between ‘ hard coded’, and ‘ hot blooded’, a place between silicon and sinew.

SMS bots are great because they can leverage the free interaction that all mobile phone users understand?—?sending a text. Leveraging the free, already learned mechanics and interactions of SMS or messaging opens a new world of possibilities for finding what you want most right now. Going through messaging apps or SMS, means you have instant access to hundreds of millions of people right out the gate, and you don’t have to educate them on using your product.

A buddy of mine is working on a bot right now, and they are doing great things for events. The bot is called Gather. Gather is your event bot. Gather is still in it’s infancy, but has already won over Startup Grind and will be debuting at the Global Conference in San Francisco this year.

The team is lead by a very capable and ambitious developer. He is already designing Gather to have personality, and feel less like a pre-programmed series of menus, and more of like, well, a friend. Right now it’s simple. Throw in an unexpected emoji here, or the perfect gif there, and you are already off to a great start. It’s not something a bot would do, it’s something a human would do.

New vs free is a simple way of pointing out that everyone understands texting. So, anything where the main interaction is texting, you get that all for free. You get customer understanding for free. Education is free. Adoption is free. Choosing another medium for the interaction is not wrong, but if it is new, or different, then you need to spend time educating your customers what it is and how to use it, then you have to convince that it’s worth doing it this way. SMS is free.

When given the choice between new and free interaction, it is almost always best to choose free.

Despite the land rush to this now widely accessible bot territory, there is still plenty of room to stand out. We have only scratched the surface of interesting things that chat bots can help improve. When building a bot, you will either try and tackle something completely new and interesting or build something very useful. It’s nice to be useful, but it’s more important to be interesting.

A new niche, a silly premise, a wild idea can make you interesting right out the gate. Which is a good thing. If you are not interesting, your bot dies. But the problem you try and solve might not be new, silly or wild, it might just be an area that makes perfect sense?—?something that would actually be very useful.

Chances are others will have the same idea, so the ‘be interesting’ mandate applies even more. I’ve seen a lot of SMS bots in preparing for this, and I’ve noticed a few things that can give life to your code.

  1. Start small?—?focusing on performing one task is key. The more narrow the niche the more interesting and thoughtful you can be, and the more fun it will be to use.
  2. Anchors and orbitals?—?split the menu system responses up into two components: anchors and orbitals. Anchors are things that always appear the same way in the message. Anchors are to maintain consistency throughout the experience from use to use. Orbitals revolve around the anchor, and might take different forms to keep the experience fresh from use to use. The anchor might be an emoji, a key word or phrase that always appears in the same spot, but hopefully the anchor is small, and the orbitals wrap around it to offer countless variations and permutations. Implementations may vary greatly here.
  3. Colloquial?—?formality in language or tone can make a bot feel mechanical, and will be much less enjoyable to use. Remember SMS or messaging is the platform for bots for more than one reason: Don’t overlook that most chat conversation is casual, formality is out of place and feels wrong.
  4. Use it or lose it?—?the last most important suggestion is to use it everyday. Is it boring to use? Is it useful? How might it be more fun? How might you get your task done sooner? Identify the places that need the most love and go back through steps 1–3 until you smile every time you use it.
It’s great to be useful, but it’s more important to be interesting.

Appless services, programmed personalities, characters as a service and smart bots are a trending up, and are a place with burgeoning opportunities. Whether you want to make the next chat bot, a slack bot for businesses, a cat that tells you the weather, or anything else you can dream up, people have never cared more about this then they do now. So, get started, and let me know what you build, I can’t wait to see.

How do you get started? Great question. Here are some tools and resources to help get going:

  1. Twilio?—?The Twilio SMS and MMS Quickstart demonstrates how to build an application that sends and receives SMS and MMS messages usingTwilio Markup Language (TwiML) and the Twilio REST API.
  2. Connecting a Bot to Twilio?—?This blog post demonstrates how to connect a Pandorabot to SMS using Twilio.

This is part 1 in a series on Characters as a service: a look into the ways we pay for personality in our products. I’ll dive deep into designing a bot with a personality, show a few case studies about the power of personality and avatars. Next up: Physical products, After that: Avatars.

Steven Perry, M.S.

Senior Electrical Engineer at VPI Technology

9 年

Great insights Mason! I look forward to part 2.

回复
Avery Duffin

Software Engineering Manager at Mastercard, Providing Risk Analysis For Issuers and Acquirers, Moonlight Developer, Budding Real Estate Investor, Husband, Father

9 年

Great thoughts Mason. I think this is exactly where the next age of innovation is coming from. Funny you write this right now. I'm actually working on a personal project right now using twilio to create a type of AI. We should talk sometime.

要查看或添加评论,请登录

Mason Wooley的更多文章

  • My New Library

    My New Library

    I added 52 new books to my library this year. Oh? You noticed? Nice.

    2 条评论
  • Using Kickstarter to fund your Kickstarter

    Using Kickstarter to fund your Kickstarter

    Ideas are like crack. I think I‘m an addict.

  • Keep Going and Never Stop

    Keep Going and Never Stop

    I am about to embark on a one year journey pursuing my crazy idea. I am writing this post so that in a few months when…

    1 条评论
  • Keep Calm and Carry on

    Keep Calm and Carry on

    About two months ago I was always stressed. I was killing it at work, moonlighting, having a baby--stuff was always…

    4 条评论
  • You have the freedom to meet your own expectations (like reading a book a week)

    You have the freedom to meet your own expectations (like reading a book a week)

    So this is what keeping your new years resolution feels like. I did it.

    3 条评论
  • Theory: Faceless Apps

    Theory: Faceless Apps

    Recently, I have been designing a few new apps all exploring faceless interfaces. Basically, that just means I am…

    1 条评论
  • Invention Debt

    Invention Debt

    I'll keep this one short. The basic idea is simple.

    1 条评论
  • Notes - Startup Grind: The best quotes from the 2016 Global Conference, and elsewhere

    Notes - Startup Grind: The best quotes from the 2016 Global Conference, and elsewhere

    I was combing over my notes and found a number of good quotes from Startup Grind 2016, so I thought I would share them.…

  • They called me Student Innovator of the Year: One Year Later

    They called me Student Innovator of the Year: One Year Later

    The whole student competition thing is a game. Games have rules, and strategies that you can learn to give you an edge…

    4 条评论
  • GO HARD, and other stories.

    GO HARD, and other stories.

    30 June 2015 - This one was big this year for me, and goes hand in hand with "making decisions out of hope instead of…

    3 条评论

社区洞察

其他会员也浏览了