When it comes to designing chatbots and voicebots, don’t over-polish your dialogue.
Dialogue designers debate the words that go into any conversational user interface.
It doesn’t matter whether you’re designing a chatbot or a voice bot, the designer, the creator behind it will deliberate over every single syllable.
Every word has to fight for its place…
And you’ll go around the houses: the design team, the client, everyone is going to have an input.
There’s personas, there’s brand personas, there’s guidelines, there’s words we do use, words we don’t use. Everything’s rehearsed.
It’s a little bit like writing a screenplay if you speak to any writers, you know, screenplay writers, they will tell you this exact thing: every word needs to fight for it’s place.
And I do agree with that, you know, but in real conversations you don’t do that.
I haven’t rehearsed these lines. I haven’t deliberated over every syllable. I’m not making every word fight for its place. I’m just talking and that’s how conversations are.
That’s how a normal, natural conversations go.
Now, I’m not saying there’s no value in deliberating every word and there’s no value in having a brand persona and all of those kind of things.
I’m not saying there’s no value in the whole team and the client debating the ins and outs of how things should sound.
What I’m asking is: what would happen if you didn’t do that?
I want to know what would happen if you didn’t deliberate over every syllable.
One of the first things we do when we design conversations is role play.
Two chairs back-to-back – socially distanced, of course, and just role play the dialogue and see what happens. See what comes out.
That forms the first pass happy path dialogue.
I’d like to know what would happen if you did that for the entire thing and you didn’t edit a line. You didn’t edit a syllable. You just took exactly, verbatim, word for word what was said in the role-play and use that exact dialogue.
I know that it is important for the brand to have their tone, their dialogue and to control that experience, but the user on the other end, do they care?
So I’d like to know what would happen if you put two conversational interfaces side-by-side; one that’s got very little to no editing done, just verbatim role-played conversations, and the other one that’s been edited and massages and tweaked to be just perfect because I don’t know whether users care and I’d like to see whether they do.
Maybe that’s how you actually design natural conversations.
Well, at least, it’s how we do it.
