文字起こし 🔗
WEBVTT
1
00:00:07.240 --> 00:00:10.539
Hello, and welcome to another episode of 600 Seconds in
2
00:00:10.779 --> 00:00:13.880
HR Tech. We are at HR Tech in Las Vegas,
3
00:00:13.980 --> 00:00:18.019
and today I'm joined by Ike Penyon. Hi. Hi.
4
00:00:18.300 --> 00:00:19.059
So good to see you again.
5
00:00:19.079 --> 00:00:22.179
Me too. So can you tell our listeners a little
6
00:00:22.219 --> 00:00:25.339
bit about you, your role at Visior, and Visior?
7
00:00:25.539 --> 00:00:28.219
Yeah, absolutely. So my name is Ike Fenyon. Great to
8
00:00:28.260 --> 00:00:30.739
meet you all virtually and then hopefully in person soon.
9
00:00:31.100 --> 00:00:32.880
So I've been a visitor for the last four years.
10
00:00:33.039 --> 00:00:36.060
I head up our product management team for our agents,
11
00:00:36.439 --> 00:00:38.159
which we'll talk about quite a bit. So our agent
12
00:00:38.179 --> 00:00:41.979
solutions for talent management, also for our platform that ingests
13
00:00:42.020 --> 00:00:44.270
and creates context. Again, another item that we'll get into
14
00:00:44.649 --> 00:00:46.689
as well as workforce planning. So how do you go
15
00:00:46.750 --> 00:00:49.750
from not just the budget level down to the layers
16
00:00:49.810 --> 00:00:52.609
below so that you can orchestrate on a position level?
17
00:00:53.240 --> 00:00:55.280
on a work, how do you shape work as it
18
00:00:55.320 --> 00:00:58.920
translates and disrupts into the future, all that sort of stuff.
19
00:00:59.159 --> 00:01:02.539
So in my past lives, I worked at Cornerstone on demand,
20
00:01:02.979 --> 00:01:06.549
overseeing portfolio strategy for Saba as well as Cornerstone Solutions,
21
00:01:06.909 --> 00:01:10.450
and then HireVue and Structure. So quite a journey through
22
00:01:10.469 --> 00:01:12.569
the HR tech landscape. So it's always been good to
23
00:01:12.909 --> 00:01:16.750
interact with you, Red Thread, as well as Danny and Stacia. Yeah, likewise.
24
00:01:16.810 --> 00:01:19.900
Very cool. So I know Vizzer does a lot, but
25
00:01:19.920 --> 00:01:22.120
today I want to talk about something that you all
26
00:01:22.340 --> 00:01:25.079
recently released. So this past September, you all released a
27
00:01:25.140 --> 00:01:28.700
connector with Microsoft Viva Glint. So can you tell our
28
00:01:28.879 --> 00:01:31.500
listeners a little bit about what that connector is, what
29
00:01:31.549 --> 00:01:34.430
it does, and why Glint? Why not any other platform?
30
00:01:34.890 --> 00:01:37.510
Yeah, it's an excellent question. Engagement data is one of
31
00:01:37.569 --> 00:01:41.230
that first underpinnings of everything else in the insight world.
32
00:01:41.609 --> 00:01:44.609
It's one of those things that most employers do with
33
00:01:44.629 --> 00:01:47.150
their customer base. And so it's the question of what
34
00:01:47.170 --> 00:01:49.090
do you actually do with it then? So great. We
35
00:01:49.109 --> 00:01:51.769
have the insights back. We see where people feel on
36
00:01:51.849 --> 00:01:55.409
certain topics, their overall sentiment, but it doesn't really point
37
00:01:55.510 --> 00:01:58.010
to what is the actual action that you should take.
38
00:01:58.489 --> 00:02:01.049
So we were very fortunate to work with Glint and
39
00:02:01.109 --> 00:02:04.579
they started to build an API to get that data out.
40
00:02:05.079 --> 00:02:07.519
And we're looking at how do they tell that value
41
00:02:07.579 --> 00:02:10.800
story to the broader market? We are a partner with Microsoft.
42
00:02:11.219 --> 00:02:13.840
And so we found a really great opportunity to help
43
00:02:13.939 --> 00:02:17.870
take that Glint engagement data, wrap it with other data
44
00:02:17.930 --> 00:02:22.250
and context from the enterprise ecosystem, then help managers to
45
00:02:22.270 --> 00:02:25.030
understand not just what was happening on a more granular level,
46
00:02:25.509 --> 00:02:27.610
but then what you should do about it. And so
47
00:02:27.669 --> 00:02:31.280
Microsoft was excited about that part to integrate Glint data
48
00:02:31.319 --> 00:02:33.159
for the first time into the broader ecosystem. We were
49
00:02:33.419 --> 00:02:36.759
excited to be a founding partner on that API, but
50
00:02:36.780 --> 00:02:39.240
then also to help our customers who had been asking
51
00:02:39.300 --> 00:02:41.639
for this type of integration to make it more streamlined.
52
00:02:42.020 --> 00:02:44.439
So you can get that insight more rapidly, more on demand.
53
00:02:44.889 --> 00:02:47.490
on a regular basis so that your managers know what
54
00:02:47.509 --> 00:02:48.430
to do with it. Yeah.
55
00:02:48.889 --> 00:02:53.699
So, okay. Can you help our listeners maybe understand this
56
00:02:53.759 --> 00:02:56.180
better by giving us an example? What do those insights
57
00:02:56.259 --> 00:02:58.719
actually look like? Where do they land?
58
00:02:58.740 --> 00:03:00.539
And who owns them at the end of the day?
59
00:03:01.139 --> 00:03:03.659
So it's a great question. And that's what Vizier ultimately
60
00:03:03.699 --> 00:03:05.340
does is one of our strengths is that we were
61
00:03:05.370 --> 00:03:07.830
built for the people space to reconcile data from across
62
00:03:07.849 --> 00:03:11.490
the ecosystems. So from your HRI system with Workday, you
63
00:03:11.509 --> 00:03:13.590
might use it for your record keeping. You don't use Pecan.
64
00:03:13.620 --> 00:03:17.259
You use Microsoft Glint. You use Greenhouse as an ATS.
65
00:03:17.280 --> 00:03:20.099
You have iSolve for benefits, et cetera, et cetera. There's
66
00:03:20.099 --> 00:03:22.780
16 on average. I think that's even maybe on the
67
00:03:22.819 --> 00:03:26.280
low side solutions that an HR department uses. And within
68
00:03:26.340 --> 00:03:28.879
each one of those is a little bit of data
69
00:03:28.919 --> 00:03:31.379
that helps to tell the story of what was done
70
00:03:31.400 --> 00:03:33.840
by who
Load full transcript
to drive the workforce as it is today.
71
00:03:34.569 --> 00:03:36.870
And HR has really struggled to actually tell the final
72
00:03:36.909 --> 00:03:39.250
story of what's the outcome to the top line, what's
73
00:03:39.280 --> 00:03:41.699
going on with the business. That's what a lot of
74
00:03:41.840 --> 00:03:44.520
practitioners are constantly gunning for. And the reason is, is
75
00:03:44.560 --> 00:03:46.740
that you don't have even all the actions into one
76
00:03:46.780 --> 00:03:48.840
place to let alone then tell the story of what
77
00:03:49.340 --> 00:03:52.439
impact those actions made. So when you take engagement data,
78
00:03:52.460 --> 00:03:55.419
for instance, now we can look at why a certain
79
00:03:55.439 --> 00:03:59.810
department is seeing much lower engagement in this most recent
80
00:03:59.889 --> 00:04:03.069
quarter of surveying, whereas the quarter before everything seemed fine.
81
00:04:03.930 --> 00:04:06.969
Turns out a driver of that might be manager retention
82
00:04:07.009 --> 00:04:10.500
and instability. So you had a prominent manager leave, high
83
00:04:10.520 --> 00:04:14.039
value talent leave. And oh, that's probably a big driver
84
00:04:14.099 --> 00:04:16.620
and reason that wouldn't show up in a question that
85
00:04:16.639 --> 00:04:18.439
you could ask your workforce. You don't ask like, are
86
00:04:18.480 --> 00:04:21.459
you unhappy because of manager instability? It comes from all
87
00:04:21.500 --> 00:04:25.250
the other data that enriches your employee engagement data to
88
00:04:25.290 --> 00:04:27.290
then point to, oh, we probably need to spend some
89
00:04:27.310 --> 00:04:30.290
time with managers, make sure everyone's happy, keep them in seat.
90
00:04:30.680 --> 00:04:34.019
So therefore, we keep engagement high and retention high in
91
00:04:34.040 --> 00:04:37.379
our workforce. And that insight only comes from creating the
92
00:04:37.420 --> 00:04:41.379
connective view, the high rich context across the enterprise data
93
00:04:41.459 --> 00:04:44.579
ecosystem that you can't do with Glint alone. You can't
94
00:04:44.620 --> 00:04:45.639
do with Workday alone.
95
00:04:46.000 --> 00:04:50.319
Right, yeah. So I know you explained to our leaders
96
00:04:50.339 --> 00:04:53.300
why bringing this data is more important now with this
97
00:04:53.360 --> 00:04:55.730
context there. But I want to specifically talk about what's
98
00:04:55.759 --> 00:04:58.250
happening now. Like, why is this so much more important
99
00:04:58.350 --> 00:04:59.089
in this moment?
100
00:04:59.329 --> 00:05:01.410
You go to any booth here. And you'll see the
101
00:05:01.449 --> 00:05:06.689
word agents, AI, take action, that sort of thing. And
102
00:05:06.990 --> 00:05:09.430
you know, my boss loves to use the phrase, all
103
00:05:09.470 --> 00:05:11.930
roads will take you to the destination if you don't
104
00:05:11.949 --> 00:05:15.069
have one. And so where are you now is an
105
00:05:15.089 --> 00:05:17.490
important point to figure out where you need to go.
106
00:05:17.930 --> 00:05:21.050
And now that we're automating and we're using AI that
107
00:05:21.110 --> 00:05:23.910
may not have the knowledge that you and I have
108
00:05:23.949 --> 00:05:27.860
about working in our individual workspaces, context becomes really important
109
00:05:28.209 --> 00:05:32.089
to make AI reliable. to make it dependable and high
110
00:05:32.149 --> 00:05:35.209
impact use cases, and also to make it governed and
111
00:05:35.250 --> 00:05:37.610
secure so that I as a manager know that the
112
00:05:37.649 --> 00:05:41.029
data that I'm using is actually high quality, it is
113
00:05:41.069 --> 00:05:43.480
the right data, and I have the right tools to
114
00:05:43.509 --> 00:05:46.259
then execute on it. And so I think the moment
115
00:05:46.319 --> 00:05:48.699
is now to make sure that enterprises not only have
116
00:05:49.160 --> 00:05:52.180
the building blocks to create context, but an ontology that
117
00:05:52.300 --> 00:05:55.220
is dependable, that is the gap for HR that just
118
00:05:55.259 --> 00:05:58.370
hasn't existed to this point. Visier, with the partnership of
119
00:05:58.430 --> 00:06:02.730
our customers, 85,000 accounts around the world, have built that
120
00:06:02.829 --> 00:06:05.290
gap for HR. And so now there is a rich
121
00:06:05.329 --> 00:06:09.009
context that AI can be more dependable, governed, and high
122
00:06:09.050 --> 00:06:13.019
value as they operate to automate decisions and to assist
123
00:06:13.060 --> 00:06:14.240
managers in the workforce.
124
00:06:14.319 --> 00:06:16.779
Yeah, so critical right now. Okay, I want to wrap
125
00:06:16.860 --> 00:06:19.220
up our conversation today by asking you about what you're
126
00:06:19.259 --> 00:06:21.790
generally seeing here. I know you've been coming to these
127
00:06:21.810 --> 00:06:22.689
kind of events.
128
00:06:22.730 --> 00:06:23.209
Long time.
129
00:06:23.879 --> 00:06:26.350
So what is it something or tell us about something
130
00:06:26.370 --> 00:06:28.750
that you're seeing different this time than you have before?
131
00:06:29.529 --> 00:06:31.569
I think that there's a new reality when it comes
132
00:06:31.589 --> 00:06:34.529
to AI. I think for a second, and it still
133
00:06:34.569 --> 00:06:36.649
pervades a little bit, right? Which everyone believes that they
134
00:06:36.689 --> 00:06:40.350
can build everything. Which on an individual perspective, if I
135
00:06:40.389 --> 00:06:43.079
want a budgeting tool just for myself or a tool
136
00:06:43.100 --> 00:06:46.040
just for my team, sure. I think that world exists.
137
00:06:46.600 --> 00:06:48.579
I think that we're a lot more sober now to
138
00:06:48.680 --> 00:06:51.220
the fact that to get the data foundation for AI together,
139
00:06:51.259 --> 00:06:54.600
to get the context for AI together, governance, security, is
140
00:06:54.800 --> 00:06:56.800
not a
trivial task and it's not one that should
141
00:06:56.839 --> 00:06:59.850
be trivial for enterprises, I think that the world is
142
00:06:59.870 --> 00:07:01.649
starting to wake up to that point a little bit more.
143
00:07:01.990 --> 00:07:06.490
So how do you use that foundation of good governance, security,
144
00:07:06.649 --> 00:07:11.620
data foundations, and build really powerful artificial intelligence and you're
145
00:07:11.740 --> 00:07:14.100
using the benefits of both? I think that's different. I
146
00:07:14.139 --> 00:07:17.759
think last year everyone was just like AI crazed. We
147
00:07:17.800 --> 00:07:19.500
had the stars in our eyes that it was going
148
00:07:19.519 --> 00:07:22.439
to fix everything. I think the tone is a little
149
00:07:22.449 --> 00:07:24.519
bit more sober, but still really optimistic in what AI
150
00:07:24.540 --> 00:07:24.879
can do.
151
00:07:24.899 --> 00:07:25.800
Yeah.
152
00:07:26.319 --> 00:07:26.560
Great.
153
00:07:26.639 --> 00:07:29.379
This has been really wonderful, Ike. Thank you so much
154
00:07:29.399 --> 00:07:31.620
for joining us in this conversation today. And thank you
155
00:07:31.639 --> 00:07:33.860
for listening in to this episode. We'll see you at
156
00:07:33.920 --> 00:07:34.480
the next one.
157
00:07:35.040 --> 00:07:38.000
So that was my conversation with Ike Bennion, Vice President
158
00:07:38.019 --> 00:07:42.209
of Product, Talent, Agents, Planning, and Platform at Vizier. I
159
00:07:42.250 --> 00:07:44.689
want to spend just a few minutes talking about what
160
00:07:44.750 --> 00:07:47.500
I took away from that conversation. And here are the
161
00:07:47.540 --> 00:07:50.060
three things that stood out to me. First, let's talk
162
00:07:50.100 --> 00:07:53.540
about the Glint connector. So Vizio released a connector with
163
00:07:53.579 --> 00:07:58.170
Microsoft Viva Glint in September. I said, scores show how
164
00:07:58.230 --> 00:08:01.350
people feel, but do not say what to do. So
165
00:08:01.430 --> 00:08:04.089
in the example that he shared in the video, a
166
00:08:04.129 --> 00:08:07.910
department's engagement scores drop. The cause may be a manager
167
00:08:07.949 --> 00:08:11.810
who left and no survey questions asks about that. Only
168
00:08:11.920 --> 00:08:15.100
data from other systems show that and so it's hard
169
00:08:15.139 --> 00:08:17.800
to know what is the right action to take. This
170
00:08:17.879 --> 00:08:21.160
fits what we see across the market. In our 2026
171
00:08:21.160 --> 00:08:25.439
People Analytics technology report, we found that 91% of vendors
172
00:08:25.639 --> 00:08:29.490
say they help leaders act on insights. Insights from survey
173
00:08:29.529 --> 00:08:32.870
data are based on limited data collected in that survey.
174
00:08:33.600 --> 00:08:36.220
So when combined with other types of data, such as
175
00:08:36.639 --> 00:08:40.500
HRIS data or ATS data or data from other systems,
176
00:08:41.000 --> 00:08:45.009
users can get much more richer insights. But vendors now
177
00:08:45.129 --> 00:08:50.029
compete with each other on action and not on insights anymore.
178
00:08:50.409 --> 00:08:52.710
So what I'd be interested in seeing is whether a
179
00:08:52.789 --> 00:08:56.450
system tracks if an action was actually taken based on
180
00:08:56.470 --> 00:08:59.639
the recommended action that was provided. And then once that
181
00:08:59.700 --> 00:09:03.029
action was taken, what happened? I'd also be interested in
182
00:09:03.090 --> 00:09:06.070
seeing if other survey providers follow this route like Lindt
183
00:09:06.190 --> 00:09:09.389
and build similar partnerships as having a holistic view of
184
00:09:09.409 --> 00:09:13.509
the data becomes more and more critical within organizations. This
185
00:09:13.629 --> 00:09:15.769
other thing that I want to talk about is context.
186
00:09:16.389 --> 00:09:20.480
Ike said HR uses about 16 tools on average and
187
00:09:20.519 --> 00:09:23.039
that he said is probably a lower number across a
188
00:09:23.460 --> 00:09:27.159
given organization. And each tool holds part of the story.
189
00:09:27.779 --> 00:09:31.159
He said, Vizier built a shared structure for HR data
190
00:09:31.340 --> 00:09:35.659
and ontology, which makes AI reliable. I agree that AI
191
00:09:35.759 --> 00:09:38.669
needs this, this holistic context that is built by connecting
192
00:09:38.710 --> 00:09:42.230
together all the different data within the organization. AI on
193
00:09:42.429 --> 00:09:45.909
unconnected people data gives answers that read well but make
194
00:09:45.970 --> 00:09:50.090
little sense. This is where human review, judgment, and expertise
195
00:09:50.149 --> 00:09:53.220
becomes critical. Humans need to have the expertise and the
196
00:09:53.269 --> 00:09:56.100
knowledge necessary to know whether an AI output is actually
197
00:09:56.139 --> 00:09:58.750
reliable or just looks good on the surface but is
198
00:09:58.789 --> 00:10:01.740
in fact inaccurate. So as a buyer, I would want
199
00:10:01.769 --> 00:10:04.379
to know where does a human check the output before
200
00:10:04.419 --> 00:10:07.379
someone acts on it. I'd also want to know how
201
00:10:07.399 --> 00:10:11.700
a system that integrates data from other systems handles when
202
00:10:11.759 --> 00:10:15.360
two systems disagree. The third thing that I want to
203
00:10:15.419 --> 00:10:18.179
talk about is the change in mood that Ike mentioned
204
00:10:18.379 --> 00:10:21.789
about the conference. Ike said that last year, many people
205
00:10:21.870 --> 00:10:25.470
believed that they could build everything with AI for a
206
00:10:25.509 --> 00:10:29.269
personal tool that still holds. But for an enterprise, the
207
00:10:29.289 --> 00:10:33.610
data foundation, context, governance and security are hard work. He
208
00:10:33.649 --> 00:10:36.629
said buyers now see this and he called the tone
209
00:10:36.669 --> 00:10:40.940
more sober and still optimistic. This speaks to the build
210
00:10:41.000 --> 00:10:44.200
versus buy debate that we've seen resurfaced over the recent past.
211
00:10:44.830 --> 00:10:48.590
When does that sell data foundation gain when buyers see
212
00:10:48.649 --> 00:10:52.419
what building actually costs in reality? It also suggests that
213
00:10:52.440 --> 00:10:54.820
the answer is not one or the other. It's not
214
00:10:54.860 --> 00:10:57.700
whether you just build or you just buy. We're now
215
00:10:57.759 --> 00:11:01.340
seeing companies buy the technology for its capabilities, but build
216
00:11:01.480 --> 00:11:04.210
their own experience layer that is added to the top
217
00:11:04.590 --> 00:11:08.129
because the experience and agent layers are now cheaper to build.
218
00:11:08.730 --> 00:11:11.350
So the question is becoming more about which layer do
219
00:11:11.370 --> 00:11:14.899
you need to control? Is it the data, the business rules,
220
00:11:14.960 --> 00:11:19.259
the employee experience, or the AI agents? So that's what
221
00:11:19.279 --> 00:11:21.960
stood out to me from our conversation with Vizier. If
222
00:11:22.000 --> 00:11:24.200
the conversation was interesting to you and you would like
223
00:11:24.240 --> 00:11:27.090
such similar insights, do check out our research. And if
224
00:11:27.110 --> 00:11:29.590
you found this useful, share it with a friend and
225
00:11:29.669 --> 00:11:32.009
share your thoughts with us. Thank you so much for listening.