1
00:00:00,300 --> 00:00:06,230
Hello and welcome back to Parity
at Rs, the retail podcast. 

2
00:00:06,290 --> 00:00:10,500
I'm joined by the CEO of 
Amperity, Barry Padgett. 

3
00:00:10,640 --> 00:00:13,800
Thank you so much for giving us 
the time to to talk to us. 

4
00:00:14,050 --> 00:00:17,340
There was a group of people that
need to usher out of the way, so

5
00:00:17,410 --> 00:00:19,260
I can imagine it's been non-stop
for. 

6
00:00:19,270 --> 00:00:20,780
You. 
It's been fantastic. 

7
00:00:21,570 --> 00:00:25,340
I can't wait to go home and 
sleep for three days for it 

8
00:00:25,350 --> 00:00:27,160
feels like same for you though, 
I'm sure. 

9
00:00:27,230 --> 00:00:28,820
Well, yeah. 
Well, we started with retail 

10
00:00:28,830 --> 00:00:32,860
safaris on Saturday and Friday, 
so we ran anyway. 

11
00:00:32,970 --> 00:00:35,080
We were at CES last week in 
Vegas. 

12
00:00:35,090 --> 00:00:36,660
Ohh like. 
In RF this week. 

13
00:00:36,670 --> 00:00:38,120
And so it show season. 
Pursue. 

14
00:00:38,130 --> 00:00:40,080
Arts. 
So listen, I think for those who

15
00:00:40,090 --> 00:00:44,240
don't know and piracy, if you 
don't mind just giving us a 

16
00:00:44,630 --> 00:00:50,250
snapshot to your journey to NRF.
And then my first question would

17
00:00:50,260 --> 00:00:53,500
be like what's been the one 
thing that retailers come to 

18
00:00:53,510 --> 00:00:55,200
ask? 
Yeah, you know, I know ask 

19
00:00:55,210 --> 00:00:57,160
Tamara, we were talking about 
retail. 

20
00:00:57,170 --> 00:01:01,540
You've actually blown away by 
the number of yeah, which tells 

21
00:01:01,550 --> 00:01:03,980
us that still running. 
So but anyway, yeah. 

22
00:01:04,190 --> 00:01:06,900
So tell me, what's the birds, he
How did you get here? 

23
00:01:07,190 --> 00:01:08,260
And what's Hall? 
Sure. 

24
00:01:08,590 --> 00:01:11,820
So we are in the data 
unification business and the 

25
00:01:11,830 --> 00:01:14,500
easiest way to describe it I 
would say is you know whether 

26
00:01:14,510 --> 00:01:18,940
you have a project that's a 
marketing project and ecom, so 

27
00:01:18,950 --> 00:01:23,220
loyalty, AI, usually you can 
trace back whether those 

28
00:01:23,230 --> 00:01:26,280
projects are successful or not 
down to the quality of your 

29
00:01:26,290 --> 00:01:29,320
customer data and your ability 
to actually unify that customer 

30
00:01:29,330 --> 00:01:31,480
data and feed it into those 
systems. 

31
00:01:31,810 --> 00:01:34,490
So the easiest way to maybe 
describe is, I think over the 

32
00:01:34,500 --> 00:01:37,620
last several years most people 
have thought about that problem 

33
00:01:37,630 --> 00:01:39,800
as a vitamin. 
Hey, it would be nice to kind of

34
00:01:39,810 --> 00:01:43,230
Turbocharge our marketing tools 
or turbocharge our paid media 

35
00:01:44,080 --> 00:01:47,010
and this does work or maybe feed
our AI projects. 

36
00:01:47,180 --> 00:01:49,890
I'd say that the pendulum has 
shifted a little bit to be less 

37
00:01:49,900 --> 00:01:53,180
vitamins and more medicine given
that a lot of those projects are

38
00:01:53,190 --> 00:01:55,580
broken now. 
And so there's a lot of things 

39
00:01:55,590 --> 00:01:59,570
between security and compliance,
cookies, AI, that's driving 

40
00:01:59,580 --> 00:02:04,490
retailers to embrace their first
party data, that data organised 

41
00:02:04,500 --> 00:02:08,080
so they can go in the server. 
I I think having voting 

42
00:02:08,090 --> 00:02:11,240
technology, I think people take 
for granted the fact that you 

43
00:02:11,250 --> 00:02:13,320
need your data to be in some 
shape. 

44
00:02:13,330 --> 00:02:15,330
Or form. 
Before you can get the output. 

45
00:02:15,390 --> 00:02:17,210
Yeah, is that right? 
Am I right? 

46
00:02:17,430 --> 00:02:18,420
Yeah. 
Or is that like? 

47
00:02:19,860 --> 00:02:21,320
No, I think you're you're spot 
on. 

48
00:02:21,330 --> 00:02:23,790
I think the difference maybe 
that we're experiencing right 

49
00:02:23,800 --> 00:02:27,740
now is most folks had a fairly 
siloed view of like I wanna get 

50
00:02:27,750 --> 00:02:30,090
my marketing data organised in 
my marketing. 

51
00:02:30,220 --> 00:02:33,230
Yeah, I'm gonna get my loyalty 
data organised in my loyalty 

52
00:02:33,240 --> 00:02:36,120
plan. 
I think that the overriding 

53
00:02:36,130 --> 00:02:39,810
demand now both for us as 
consumers as well as with the 

54
00:02:39,820 --> 00:02:43,000
systems we're deploying, there 
has to be interoperability, 

55
00:02:43,110 --> 00:02:46,120
there has to be compatibility 
between those datasets. 

56
00:02:46,270 --> 00:02:49,260
And that's I think where we're 
seeing everyone state to now is 

57
00:02:49,510 --> 00:02:54,200
to unify this data, eat and 
hydrate all my systems with one 

58
00:02:54,210 --> 00:02:57,200
single view of the customer, 
whether it's because of again 

59
00:02:57,370 --> 00:03:01,000
control or compliance, GDPR in 
Europe, CDA out weren't you? 

60
00:03:01,050 --> 00:03:02,920
Yeah. 
Or because I'm trying to train a

61
00:03:02,930 --> 00:03:06,680
large language model AI 
projects, I have to get the data

62
00:03:06,690 --> 00:03:09,080
right. 
But those AI projects really 

63
00:03:09,090 --> 00:03:11,990
amplify whether you have good 
quality data or you. 

64
00:03:12,000 --> 00:03:14,600
Don't. 
Yeah, so I'm a retailer. 

65
00:03:14,610 --> 00:03:19,150
I've invested heavily in trying 
to solve this problem and it's 

66
00:03:19,200 --> 00:03:21,340
it's going horrible. 
Where do I start? 

67
00:03:21,350 --> 00:03:24,670
Yeah, because what I often see 
is this sort of well and Bush 

68
00:03:24,720 --> 00:03:28,650
from imagine finance from 
marketing merchandising. 

69
00:03:28,660 --> 00:03:29,470
Yeah. 
And pension. 

70
00:03:29,480 --> 00:03:31,350
It feels like technology is 
trying to solve. 

71
00:03:31,860 --> 00:03:33,870
Yeah, right. 
The tech they don't really 

72
00:03:33,880 --> 00:03:36,380
understand merchandisers data 
needs are. 

73
00:03:36,390 --> 00:03:38,230
Different from. 
The ERP. 

74
00:03:38,240 --> 00:03:40,890
System that's curious. 
How did you help? 

75
00:03:41,870 --> 00:03:44,790
Yeah, solving. 
Well, it's it's very similar to 

76
00:03:44,800 --> 00:03:47,360
what we've been thinking about 
the CRM category. 

77
00:03:47,370 --> 00:03:50,940
You know, 20 years ago it was 
cool to build your own CRM. 

78
00:03:51,730 --> 00:03:56,460
That would be a tough project. 
Yeah, it would be hard to pitch 

79
00:03:56,470 --> 00:04:01,590
a build it yourself CRM project.
I feel like this customer data 

80
00:04:01,720 --> 00:04:05,130
and the customer data category 
is also purely new in the sense 

81
00:04:05,140 --> 00:04:08,730
that most retailers have been 
trying to build this unified 

82
00:04:08,740 --> 00:04:12,590
view for a long, long time using
lots of tools, ETL and the dupe 

83
00:04:12,600 --> 00:04:15,230
data lakes. 
And so I feel like the 

84
00:04:15,240 --> 00:04:21,149
overwhelming sentiment now is 
that there's a, there's a data 

85
00:04:21,160 --> 00:04:25,530
supply chain count in that we 
can't get that data supply chain

86
00:04:26,120 --> 00:04:28,770
unified and built. 
And so a lot of companies are 

87
00:04:28,780 --> 00:04:32,670
now looking to off the shelf 
providers like in Parity will 

88
00:04:32,680 --> 00:04:34,370
come in. 
We have a great sort of set of 

89
00:04:34,380 --> 00:04:37,150
models that are all machine 
learning and AI based and it 

90
00:04:37,160 --> 00:04:40,990
takes the pressure off of the 
retailer through to those teams 

91
00:04:41,060 --> 00:04:43,710
and building a foundation. 
So they can use that foundation 

92
00:04:43,720 --> 00:04:45,820
to go do all the cool things 
they want to do like they'll 

93
00:04:45,830 --> 00:04:49,190
propensity models, density, 
sure, customer lifetime are you 

94
00:04:49,240 --> 00:04:52,410
eat all their downstream tools 
serve their internal customers. 

95
00:04:52,500 --> 00:04:56,060
And so I feel like no shame on 
companies trying to build this 

96
00:04:56,070 --> 00:04:58,250
themselves. 
We're just now reaching the way 

97
00:04:58,260 --> 00:05:00,510
where there's actually tooling 
out there to go help. 

98
00:05:00,580 --> 00:05:02,910
Whether you want to do that 
directly with a company like 

99
00:05:02,920 --> 00:05:06,390
Impurity, yeah, or use amperity 
tech within a lake house 

100
00:05:06,400 --> 00:05:09,170
architecture like Databricks or 
Snowflake, these are the kinds 

101
00:05:09,180 --> 00:05:12,380
of things that are are now 
available to tech teams, 

102
00:05:12,540 --> 00:05:13,950
retailers for the very first 
time. 

103
00:05:15,430 --> 00:05:18,850
Those two questions, yeah, is 
something that surprised you as 

104
00:05:18,860 --> 00:05:22,340
an outcome from one of your 
customers that has sort of gone 

105
00:05:22,350 --> 00:05:24,120
on the journey with. 
Yeah, and then? 

106
00:05:25,590 --> 00:05:26,940
Yeah, and. 
Then the second one. 

107
00:05:27,010 --> 00:05:29,540
I'd love your vision of future. 
Where do you think? 

108
00:05:29,610 --> 00:05:31,580
With AI and everything? 
What do you think we're going? 

109
00:05:31,590 --> 00:05:34,100
Yeah. 
So on the first question, say a 

110
00:05:34,110 --> 00:05:37,420
surprise for me, Brooks running,
who's here at the show 

111
00:05:37,430 --> 00:05:39,280
somewhere. 
They're a great customer of Ann 

112
00:05:39,290 --> 00:05:40,760
Parity. 
They're based in Seattle. 

113
00:05:40,770 --> 00:05:42,280
We happen to be based in Seattle
too. 

114
00:05:42,320 --> 00:05:45,470
We have a lot of great Seattle 
brands, Alaska Airlines, sports 

115
00:05:45,480 --> 00:05:48,800
from T-Mobile, Brooks running. 
One of the things I love most 

116
00:05:48,810 --> 00:05:51,880
about that story was they did 
the data unification project 

117
00:05:52,090 --> 00:05:55,730
they to sort of retention 
programme for existing customers

118
00:05:55,820 --> 00:05:59,530
new acquisition eating all their
downstream tools, my favourites 

119
00:05:59,540 --> 00:06:03,250
that bit was we got a blurb back
from Melanie Allen who's the CMO

120
00:06:03,260 --> 00:06:06,850
there and I wanted to share this
with you which is well customer 

121
00:06:06,860 --> 00:06:08,890
service reps. 
So people that are calling 

122
00:06:08,900 --> 00:06:11,650
support for some reason are 
return or refund. 

123
00:06:11,920 --> 00:06:15,780
And the customer service quote 
was with with what's going on 

124
00:06:15,790 --> 00:06:19,570
with our data now and able to 
close pieces or the customer can

125
00:06:19,580 --> 00:06:21,270
fully explain what their 
problem. 

126
00:06:22,060 --> 00:06:25,170
And the reason is when someone's
calling customer service. 

127
00:06:25,240 --> 00:06:28,540
Now that customer service reps 
like Ohh Alex is calling and it 

128
00:06:28,550 --> 00:06:31,970
looks like he bought 2 things. 
You returned one on Monday, 

129
00:06:32,260 --> 00:06:34,470
Wednesday is probably calling to
see where the refund is. 

130
00:06:34,540 --> 00:06:35,670
Yeah. 
So when you call us, yes. 

131
00:06:35,680 --> 00:06:37,750
Thanks for being a customer. 
It looks like you bought 2 pairs

132
00:06:37,760 --> 00:06:40,300
of shoes, your return one. 
Do you have a question on size 

133
00:06:40,310 --> 00:06:42,730
or you asking about the refund 
and you're like, yeah, that's 

134
00:06:42,740 --> 00:06:45,870
why I'm falling and so 
unintended, unintended to get 

135
00:06:45,880 --> 00:06:48,890
that customer service quote. 
But the fact that you can drive 

136
00:06:48,900 --> 00:06:52,150
internal employee experience is 
given that our employees want to

137
00:06:52,160 --> 00:06:54,970
serve the customer, they 
desperately want to make great 

138
00:06:54,980 --> 00:06:58,050
customer impression and giving 
them the data and the tools you 

139
00:06:58,060 --> 00:07:01,060
don't do that is fantastic, so. 
For so long. 

140
00:07:01,070 --> 00:07:02,840
Totally. 
And then your second question in

141
00:07:02,850 --> 00:07:05,440
terms of where we're going. 
I think what you're going to 

142
00:07:05,450 --> 00:07:08,140
find at least in our perspective
in the customer data space, 

143
00:07:08,330 --> 00:07:12,720
you're going to find most 
companies have really nicely to 

144
00:07:12,730 --> 00:07:15,880
large language model that's 
bespoke to their customers, 

145
00:07:16,130 --> 00:07:19,820
their customer tooling ecosystem
of technology they use, their 

146
00:07:19,830 --> 00:07:22,530
semantics and preferences. 
And you'll find that you'll be 

147
00:07:22,540 --> 00:07:26,510
able to plug in all these off 
the shelf Pi tools you know into

148
00:07:26,520 --> 00:07:29,550
your large language model and 
really drive personalization. 

149
00:07:29,760 --> 00:07:33,990
I know it's been a a a goal for 
so long personalization at 

150
00:07:34,000 --> 00:07:35,630
scale. 
Now listen like putting 

151
00:07:35,640 --> 00:07:38,930
someone's correct person name in
an outbound e-mail is not 

152
00:07:38,940 --> 00:07:41,310
personalization. 
Understanding that you're the 

153
00:07:41,320 --> 00:07:43,510
head of a household and you buy 
for your kids and you they for a

154
00:07:43,520 --> 00:07:46,630
spouse and you buy for yourself.
And really understanding what 

155
00:07:46,640 --> 00:07:49,370
that sentiment is and your goal 
and purpose for the next 

156
00:07:49,380 --> 00:07:51,490
engagement is, that's where AI 
is going to be. 

157
00:07:51,500 --> 00:07:53,330
Great. 
Takes a large language model. 

158
00:07:53,340 --> 00:07:56,460
Tune to your data and you have 
that customer data built that 

159
00:07:56,470 --> 00:07:58,310
foundation unified before you 
can trade it. 

160
00:07:58,480 --> 00:07:59,810
It was actually one of the 
talks. 

161
00:07:59,940 --> 00:08:02,870
When someone signs up to your 
loyalty programme and makes 

162
00:08:02,880 --> 00:08:06,170
themselves digitally pressed for
you, they want more. 

163
00:08:06,270 --> 00:08:08,950
That's right, I've already 
agreed all the GDPR things 

164
00:08:09,060 --> 00:08:12,260
you've been. 
Anyway, I carry on soldier, 

165
00:08:12,270 --> 00:08:14,700
here, Thank you so much for 
giving me your time. 

166
00:08:14,710 --> 00:08:18,620
And of course you are because 
and I look forward to seeing you

167
00:08:18,630 --> 00:08:19,010
guys. 
Great. 

168
00:08:19,630 --> 00:08:21,640
How are you internationally? 
Like are you? 

169
00:08:21,710 --> 00:08:23,460
Officially. 
US based their new. 

170
00:08:23,770 --> 00:08:26,680
Offices in Melbourne, Australia 
and get served by these and in 

171
00:08:26,690 --> 00:08:29,900
each pack regions from there 
opposite in London that serves 

172
00:08:29,910 --> 00:08:31,400
how you're. 
OK, yes. 

173
00:08:31,510 --> 00:08:32,429
Yeah. 
Yeah. 

174
00:08:32,440 --> 00:08:33,020
Thanks, Alex.