1
00:00:00,110 --> 00:00:06,710
Guy: Nine. Let users undo their mistakes.
Welcome to the Service Design Principles on Guy

2
00:00:06,710 --> 00:00:10,910
Martin, joined by the author of the Service
Design Principle series of books. Founder of the

3
00:00:10,910 --> 00:00:15,530
Swiss Innovation Academy and Service Design
Practitioner The error corrected Daniela.

4
00:00:15,540 --> 00:00:16,260
Caitlin Otto

5
00:00:17,000 --> 00:00:18,350
Daniele: had such a pleasure.

6
00:00:18,720 --> 00:00:24,120
Guy: Hi. So each episode, as you know, we look
at one of the principles from your book, Service

7
00:00:24,120 --> 00:00:29,310
Design Principles. One, two, one. And today it's
principle number nine. Let users undo their

8
00:00:29,310 --> 00:00:36,090
mistakes. So a lot of the work that we do in in
service design and customer experience and

9
00:00:36,090 --> 00:00:41,340
things like these is about reducing friction.
But here you're suggesting that sometimes adding

10
00:00:41,340 --> 00:00:44,670
friction could be useful. So when does it make
sense to add friction?

11
00:00:44,965 --> 00:00:51,535
Daniele: So a good example is when people are
about to delete something which cannot be

12
00:00:51,535 --> 00:00:58,165
retrieved. You know, this is a very good example
where we basically tell people, Are you really

13
00:00:58,165 --> 00:01:04,525
sure you want to do this? You know, in a perfect
world where we want to smooth every experience,

14
00:01:04,525 --> 00:01:09,325
you know, whenever you would click delete on
your computer, it would just take it out, remove

15
00:01:09,325 --> 00:01:15,475
it completely, not even put it in the trash. But
no, we add friction because we know that

16
00:01:16,015 --> 00:01:21,835
sometimes we do things a little bit quickly and
that's a good moment to do that. Another good

17
00:01:21,835 --> 00:01:29,425
example is I had this one once at LinkedIn. I
was I don't know why. In a comment I wrote

18
00:01:29,425 --> 00:01:36,535
something about Showers and LinkedIn before I, I
wanted to click to send and then it said,

19
00:01:37,975 --> 00:01:45,565
maybe just have a second thought about your post
because we're trying to have conversations about

20
00:01:45,565 --> 00:01:54,355
professional stuff. And it was like, Oh, let me
double check the did I write something that is

21
00:01:54,595 --> 00:01:58,885
that that shouldn't be on a professional network?
Then it was like, Oh no, it's good. It was just

22
00:01:58,885 --> 00:02:04,915
an image that I was giving with the word shower.
But, but they had kind of this mechanism which

23
00:02:04,915 --> 00:02:11,515
maybe showed looked for words like please shower
naked, that this kind of stuff. And just not

24
00:02:11,605 --> 00:02:19,165
saying you can't speak about that, but rather
Are you sure? Are you sure this is a good moment?

25
00:02:19,285 --> 00:02:21,655
Are you sure this is the right thing to do at
that moment?

26
00:02:22,170 --> 00:02:27,750
Guy: Right. But we don't want to be asked all
the time. So it's contextual, right? That. That

27
00:02:27,750 --> 00:02:33,390
this. This happens. Because if every decision,
every click that we do, we're asked to confirm

28
00:02:33,390 --> 00:02:35,460
that. That it would drive people insane.

29
00:02:36,325 --> 00:02:41,605
Daniele: And even it would drive another thing
which you might remember that from Windows Vista

30
00:02:41,605 --> 00:02:47,185
back in the days where it asked so many times
your permission, but people got into a mode

31
00:02:47,185 --> 00:02:52,495
whenever there was a window, they just clicked,
okay, you know, and then they don't even read it.

32
00:02:52,495 --> 00:02:57,835
So it's as you say, it's a good balance of
knowing when is the good moment and having not

33
00:02:57,835 --> 00:03:02,965
too many of those. And that's where the art of
the resistance starts. You know, it's like there

34
00:03:02,965 --> 00:03:12,775
is no perfect formula to know how much is enough
and you have to kind of be be empathetic of the

35
00:03:12,775 --> 00:03:13,375
situation.

36
00:03:13,740 --> 00:03:20,550
Guy: Right. So one thing I noticed in some email
clients is if I use the word attached or

37
00:03:20,550 --> 00:03:25,920
attachment, then and I don't attach something,
then it will pop up and say, Oh, did you mean to

38
00:03:25,920 --> 00:03:31,020
attach a file because you've said this right? So
it's saying, Here's something which I noticed

39
00:03:31,020 --> 00:03:35,730
about the content of your message that you're
saying there's an attachment, but there is no

40
00:03:35,730 --> 00:03:40,110
attachment. So are you sure you want to send
this without an attachment? But if I didn't put

41
00:03:40,110 --> 00:03:43,920
those words into the text, then it wouldn't ask
me that question.

42
00:03:44,995 --> 00:03:49,195
Daniele: And that's, you know, in a way, it goes
even further than than the principle that we

43
00:03:49,215 --> 00:03:54,115
just speaking about, which is let's use and do
mistake, which is prevent people from doing the

44
00:03:54,115 --> 00:03:54,685
mistake,

45
00:03:54,660 --> 00:03:55,050
Guy: Right.

46
00:03:54,685 --> 00:03:54,955
Daniele: you know.

47
00:03:55,080 --> 00:03:55,470
Guy: Yeah.

48
00:03:55,495 --> 00:03:59,035
Daniele: Right. And, and I think there are kind
of these two levels, you know, there is level

49
00:03:59,035 --> 00:04:05,695
one, which is making sure that people can even
do stuff. And this is something that especially

50
00:04:05,695 --> 00:04:10,795
in the digital world, you know, you have
services where if you, for example, you read the

51
00:04:10,795 --> 00:04:18,415
message, you can say, Oh, I read it. There are
there are services where until I think one year

52
00:04:18,415 --> 00:04:24,265
ago Apple Messages was like that, you know, and
I this is was very disturbing because you just

53
00:04:24,265 --> 00:04:29,515
opened the message by by mistake and you were
like, okay, now I will I will forget to come

54
00:04:29,515 --> 00:04:35,185
back to it. And so for digital services, this is
kind of like the first level is having a way to

55
00:04:35,185 --> 00:04:41,305
bring it back. Even if you delete something,
maybe adding a few days of delay before you

56
00:04:41,305 --> 00:04:46,165
delete it completely. You know, Facebook does
that with your accounts. When you delete your

57
00:04:46,165 --> 00:04:50,935
account, they're telling you, okay, we're going
to delete it in 24 hours or in so many hours.

58
00:04:51,295 --> 00:04:55,855
Until then, you still can change your opinion.
And so that's kind of like the first level. And

59
00:04:55,855 --> 00:05:01,105
the second level, which you're talking about is
a bit more. So again, now that we've done that,

60
00:05:02,035 --> 00:05:05,935
how can we prevent people from doing that kind
of mistake?

61
00:05:06,240 --> 00:05:10,500
Guy: So it's not just undoing it? Yeah. It's the
preventative is better than the cure.

62
00:05:10,795 --> 00:05:12,835
Daniele: I think both is kind of like the sweet
spot.

63
00:05:13,230 --> 00:05:17,700
Guy: Because I guess, you know, the there's.
There's economics and where people are behaving

64
00:05:17,700 --> 00:05:21,510
rationally. And then there's behavioral
economics where people are behaving irrationally

65
00:05:21,780 --> 00:05:29,340
and people make mistakes. But I guess sometimes
we design for the rational human as opposed to

66
00:05:29,340 --> 00:05:34,499
the real human, which is is sometimes irrational.
And some will make mistakes and So providing

67
00:05:34,499 --> 00:05:39,809
that way to to back out of what they've done or,
you know, corrects that mistake, I think that

68
00:05:40,169 --> 00:05:41,429
it's the important thing. Y.

69
00:05:41,894 --> 00:05:45,554
Daniele: And if we're speaking of behavioral
economics, I think two terms that people will

70
00:05:45,554 --> 00:05:51,434
love to Google are one which is nudge and the
other one which is rational, rational override.

71
00:05:51,704 --> 00:05:58,154
And these are basically two terms now just
basically saying making things more users. So

72
00:05:58,154 --> 00:06:05,174
that stuff happens and rational override is
basically saying putting a stop so that people

73
00:06:05,354 --> 00:06:10,844
can't do something because before that they
confirm it. For example, when you delete the

74
00:06:10,844 --> 00:06:16,984
file, you have a rational override which is
blocking you from what you were doing and they

75
00:06:17,024 --> 00:06:22,754
pay attention. Attention, You need to do
something. And the nudge is, for example, you

76
00:06:22,754 --> 00:06:31,144
know, when I put the cookies not in the kitchen,
but I put them in the cellar, but put some

77
00:06:31,484 --> 00:06:38,144
apples on our kitchen table, this is a nudge.
It's something where automatically I will go get

78
00:06:38,264 --> 00:06:41,204
the apples. But even without thinking about it.

79
00:06:41,299 --> 00:06:46,253
Guy: Right. Which kind of ties back to making it
easy for you to do the right thing.

80
00:06:46,028 --> 00:06:50,978
Daniele: Yeah. And so that's and so that's kind
of like the arts, like in the academic stuff. So

81
00:06:50,978 --> 00:06:57,728
if you want to go into, into the more academy
stuff, it's like basically we need to know when

82
00:06:57,728 --> 00:07:03,608
is the good time to nudge people towards
something and when is a good time to create

83
00:07:03,608 --> 00:07:10,268
rational overrides to block people. And that's
where it's like where science meets the art.

84
00:07:11,753 --> 00:07:18,093
Guy: So this particular principle is a little
bit about both. And. And that rational override.

85
00:07:18,383 --> 00:07:24,353
It's just the pause. It's allowing the the human
brain, the rational brain to take over a little

86
00:07:24,353 --> 00:07:30,083
bit and say, Hang on, monkey. Just wait. Hang on,
Mr. Lizard Brain. Just. Just wait a second,

87
00:07:30,113 --> 00:07:35,633
because this could be important. Don't just go
by instinct here. That could lead you astray.

88
00:07:36,258 --> 00:07:41,718
Daniele: Yeah. So if we wanted to say it a bit
more like academics, it would be use rational

89
00:07:41,718 --> 00:07:46,470
overrides sometimes because, you know, academics
always add sometimes maybe...

90
00:07:46,871 --> 00:07:47,661
Guy: "Depends. ". Yeah.

91
00:07:48,690 --> 00:07:50,190
Daniele: it depends on the context

92
00:07:51,861 --> 00:07:53,121
Guy: All right, great. Thank you very much.

93
00:07:53,037 --> 00:07:53,547
Daniele: Thanks to you

94
00:07:53,827 --> 00:07:54,577
Guy: See you on the next one.