192: Will Data Analysts Survive 2026? 3 Major Predictions

Speaker: 00:00:00

Avery Smith-1: 2026 is here, and here

are my three predictions of what you

2

: 00:00:04

can expect as a data analyst this year.

3

: 00:00:07

Number one, I think it's going to

actually become easier to land a

4

: 00:00:11

day job in 2026 than it was in 2025.

5

: 00:00:15

Over the last few years, there has been a

lot of false information, misinformation,

6

: 00:00:20

and a lot of confusion about what's

actually going to happen with data jobs.

7

: 00:00:24

Now, I can't say that I'm a magic

fortune teller, but I have been able

8

: 00:00:27

to look at some of the data since 2019.

9

: 00:00:30

Uh, and obviously like data

analytics was really hot from like,

10

: 00:00:33

what, 2015 to maybe 20 21, 20 22.

11

: 00:00:38

Around 2022.

12

: 00:00:39

Something crazy happened where we

maybe got a little bit saturated.

13

: 00:00:43

Um, and it's not that data jobs

went down, it's just that they kind

14

: 00:00:47

of started staying about the same.

15

: 00:00:48

From 2022 to 2025.

16

: 00:00:51

There wasn't a whole lot of growth.

17

: 00:00:54

There wasn't a whole lot of decay, but it

was kind of just stagnant where it was.

18

: 00:00:58

Uh, with that, I still think that the

data analytics and the data analyst

19

: 00:01:01

profession was still being quite hyped.

20

: 00:01:03

I mean, I understand why it is a

really awesome career, but I think

21

: 00:01:07

we've seen a lot of the hype die down.

22

: 00:01:10

I think a lot of the hype has moved

towards like, uh, AI and automation.

23

: 00:01:14

And with that I think there's

people who are probably less

24

: 00:01:16

interested in becoming an analyst.

25

: 00:01:18

Data analyst and more

interested in becoming like an

26

: 00:01:21

AI person or an AI engineer.

27

: 00:01:23

I don't even know what the

titles are for these AI roles.

28

: 00:01:25

I don't think anyone really

knows what the titles are.

29

: 00:01:27

Uh, but I think a lot of people are

less interested in AI or a lot of

30

: 00:01:31

people are less interested in data and

more interested in AI and automation.

31

: 00:01:35

And because of that, I think

you're gonna see less people

32

: 00:01:37

applying for data analyst roles.

33

: 00:01:39

Now I think this, there'll be

like the same number of data jobs

34

: 00:01:43

open in 2026 as there was in 2025.

35

: 00:01:46

But I think there's just

gonna be less competition.

36

: 00:01:48

I think people are gonna try to

get into AI and automation instead.

37

: 00:01:52

I think that's great.

38

: 00:01:52

I think AI is really cool.

39

: 00:01:54

I think automation's really cool.

40

: 00:01:55

I use both in my business.

41

: 00:01:57

Um, but you still can't beat the

bread and butter of data analytics.

42

: 00:02:01

Uh, AI is definitely really cool,

but it's also a little bit overhyped

43

: 00:02:06

and we are for sure towards the

end of some sort of AI bubble.

44

: 00:02:10

Now, once again, I'm not a fortune teller.

45

: 00:02:12

I don't know when the bubble's gonna pop,

but the bubble's gonna pop eventually.

46

: 00:02:15

Um.

47

: 00:02:16

That's, that's not to say that

I still wouldn't buy AI stock.

48

: 00:02:19

I think AI is going to

be huge down the road.

49

: 00:02:22

Um, but data analytics is a lot

more proven than AI at this point,

50

: 00:02:25

and I think it's a really good

investment for you and your career.

51

: 00:02:29

Um, it's still going to be hard to

land a data job, but I think there'll

52

: 00:02:33

be less competition next year.

53

: 00:02:35

So I think it'll be easier for people to

pivot into data analytics just because

54

: 00:02:39

it's not as hyped as it once was.

55

: 00:02:41

There'll be less people kind of applying

for those entry level, uh, data roles.

56

: 00:02:45

Uh, and I think it'll just

be a little bit easier.

57

: 00:02:48

My prediction number two is that

companies will start to adopt AI

58

: 00:02:52

more, uh, to do data analytics.

59

: 00:02:54

And that doesn't mean that

there's gonna be less jobs.

60

: 00:02:57

That doesn't mean that AI

is coming for your job.

61

: 00:02:59

It doesn't mean that it's all over.

62

: 00:03:00

Uh, data analytics is here to stay.

63

: 00:03:02

Now will it change down the road?

64

: 00:03:04

Sure.

65

: 00:03:05

I'm sure it will.

66

: 00:03:06

But like what industry hasn't changed

in like a 10 year period, right?

67

: 00:03:10

Like is the automotive

industry today the same?

68

: 00:03:12

It was 10 years ago.

69

: 00:03:13

We're still driving cars, but it looks

completely different Ev self-driving.

70

: 00:03:17

I can't even tell you like

how different it looks.

71

: 00:03:20

Every industry changes in a

decade's time, and that'll be

72

: 00:03:23

true for data analytics as well.

73

: 00:03:24

I mean, 10 years ago we didn't even have

Power bi, so we we even ignoring all

74

: 00:03:29

of the AI stuff, like data analytics is

obvious, obviously changed a lot because

75

: 00:03:33

one of the most fundamental tools, data

analytics, did not exist a decade ago.

76

: 00:03:37

I think companies are.

77

: 00:03:39

Pretty slow to adopt new technology.

78

: 00:03:41

At least like the enterprises, like

we're talking like the Fortune 500.

79

: 00:03:44

Of course there's companies

that are outliers that are gonna

80

: 00:03:48

perform well, uh, using ai.

81

: 00:03:50

Um, but a lot of companies

are slow to adopt technology.

82

: 00:03:54

They're slow to actually

implement technology.

83

: 00:03:56

And I know, 'cause I literally worked for

what, like the seventh biggest company in

84

: 00:04:00

the world at the time when I was there, I

worked for ExxonMobil as a data scientist.

85

: 00:04:04

I can't even tell you how much of their

analysis at ExxonMobil was done in Excel.

86

: 00:04:09

I'll say that again.

87

: 00:04:10

Like a lot of our analysis at

ExxonMobil was done in Excel.

88

: 00:04:16

Python's been around for how many years?

89

: 00:04:18

What?

90

: 00:04:19

1990.

91

: 00:04:19

So 35 years.

92

: 00:04:21

And we weren't even using

a ton at ExxonMobil.

93

: 00:04:25

Uh, is Is Python better than Excel?

94

: 00:04:27

In my opinion, yeah.

95

: 00:04:28

It's great, but it's hard to actually

make progress in these big companies.

96

: 00:04:34

It's hard to adopt new technologies.

97

: 00:04:36

It's hard to roll out new technologies.

98

: 00:04:38

There's all sorts of

different problems and issues.

99

: 00:04:40

Like even getting Python on

your computer at ExxonMobil was

100

: 00:04:44

probably a two week process.

101

: 00:04:46

It probably takes me, if I were to like

get a computer, it maybe takes me 30

102

: 00:04:50

minutes to get Python installed on it.

103

: 00:04:52

Right?

104

: 00:04:52

At ExxonMobil, it was like a two

to three week period, just because

105

: 00:04:55

you had to ask for permission.

106

: 00:04:57

They do all these security checks, you

had to download it, it would break.

107

: 00:05:00

It was so hard to even download Python.

108

: 00:05:03

Uh, and so these larger institutions

like Humana, Wells Fargo, chase Bank.

109

: 00:05:08

Like, I'm sure they're gonna want

to adopt ai, but it, it's going to

110

: 00:05:11

happen over years, if not decades,

where that rollout actually comes out.

111

: 00:05:16

Now, I do think a lot of enterprise

companies are going to make some progress

112

: 00:05:20

on that this year, and I think mainly it's

going to be because of the integrations

113

: 00:05:24

with the companies that are already using.

114

: 00:05:26

So, for example, a lot of enterprises have

a pretty good relationship with Microsoft.

115

: 00:05:29

They're paying for enter

enterprise, Microsoft, uh, plans.

116

: 00:05:33

And I think they're gonna do a

good job with copilot and kind

117

: 00:05:35

of mingling that with chat GPT.

118

: 00:05:37

So I think that will probably

be something that you see these

119

: 00:05:41

enterprises doing over the next year.

120

: 00:05:43

Uh, I think Google's made a lot

of progress with their AI products

121

: 00:05:46

in the last, like quarter alone.

122

: 00:05:47

So those who have a.

123

: 00:05:48

Google Enterprise plans will probably

start to use AI a little bit more,

124

: 00:05:52

but I think there's a lot of stumbling

blocks for enterprises to use AI

125

: 00:05:56

that has been existing in the past.

126

: 00:05:57

I think that'll, uh, become a

little bit less of a barrier this

127

: 00:06:01

year, uh, but still a barrier.

128

: 00:06:02

Nonetheless.

129

: 00:06:04

The way I predict this actually,

like rolling out to companies,

130

: 00:06:06

the way I see it is it'll probably

be at an individual level.

131

: 00:06:10

So a lot of like data scientists

don't even really have a

132

: 00:06:13

corporate AI plan right now.

133

: 00:06:15

Um, but I see a lot of

that changing this year.

134

: 00:06:19

There's a lot of solutions that

have made a lot of progress.

135

: 00:06:21

You guys have seen me do

sponsorships with Julius ai.

136

: 00:06:24

Um, they've made a lot of

progress with their connectors.

137

: 00:06:26

The, the biggest thing is it's

really hard to have secure and

138

: 00:06:29

connect connected data, and so Julius

has made a lot of progress there.

139

: 00:06:33

I think Hex has a, a really good

product that will make some progress.

140

: 00:06:37

Like I said, Chae, Claude and Gemini

from Google have all made a lot of

141

: 00:06:41

progress in the last little bit.

142

: 00:06:42

That makes it easier to connect to

your data and have your data be secure.

143

: 00:06:46

So I think a lot of like individual

data analysts and data scientists

144

: 00:06:50

will start to get access to ai.

145

: 00:06:53

Augmented tools, and I don't think

it's gonna be replacing them.

146

: 00:06:55

Like it's literally just a

tool for them to be using.

147

: 00:06:59

And if you think that AI's going to

replace you, to me it kind of shows

148

: 00:07:04

you haven't really used AI to analyze

data yet because it's not there.

149

: 00:07:10

It's definitely not there yet.

150

: 00:07:11

Um, and like for me the other day,

uh, I, I was analyzing some data

151

: 00:07:16

and I was just using AI to do it.

152

: 00:07:18

And like, you still have to think.

153

: 00:07:20

There's so much thinking,

there's so much planning.

154

: 00:07:22

You have to know what to do.

155

: 00:07:23

You have to have the idea.

156

: 00:07:25

You know, AI can spit out 10

ideas, but like seven of 'em

157

: 00:07:28

are usually really stupid.

158

: 00:07:30

Three of 'em you can't even do.

159

: 00:07:31

So 10 outta 10 ideas like don't even work.

160

: 00:07:34

Uh, so they still need you to, to

be thinking, um, you're going to be

161

: 00:07:37

also more of the bridge between the

analysis and the actual business.

162

: 00:07:42

We were at ExxonMobil, we automated

a lot of stuff that humans were doing

163

: 00:07:46

using Python and machine learning.

164

: 00:07:48

You think, you think that just magically

the people who were doing their job

165

: 00:07:51

lost their job and just got laid off?

166

: 00:07:53

No, that's not what happened.

167

: 00:07:55

It was just a tool to help

them do their job better.

168

: 00:07:57

And a lot of the times they actually

overruled our decisions, our, our

169

: 00:08:01

decisions as in the algorithms decisions.

170

: 00:08:03

Um, this was for buying crude oils,

like deciding what crude oils from

171

: 00:08:07

around the, the world we were gonna buy.

172

: 00:08:08

This was for deciding how much.

173

: 00:08:10

Gasoline we should send to

your local Exxon gas station.

174

: 00:08:13

Like I created a machine learning

algorithm that would basically

175

: 00:08:16

predict that, and I thought it

was pretty good, but a lot of

176

: 00:08:18

the times it was missing context.

177

: 00:08:20

A lot of the times, uh, like

these traders knew best.

178

: 00:08:23

And I think that's still

gonna be true today.

179

: 00:08:25

Like, AI is really smart, but

it's not replacing a human.

180

: 00:08:29

And, and if it is, then why?

181

: 00:08:31

Like, if it is, like why has, has

it like it's just not good enough.

182

: 00:08:36

I have tried AI to make

social media content.

183

: 00:08:39

To do data analytics, to make video

scripts, to make thumbnails, and it's

184

: 00:08:43

helpful, but it's never, ever, ever

gotten it right on the first time.

185

: 00:08:47

So I don't think AI is coming

for your job, but I do think that

186

: 00:08:51

companies will start to use ai.

187

: 00:08:52

I think your job as a data

analyst is gonna change.

188

: 00:08:56

More into that connector from the

actual data analysis to the business,

189

: 00:09:00

I think it's gonna be more important

to know what to do versus how to do it.

190

: 00:09:05

So like for example, like you can

do a pivot table in Excel, you can

191

: 00:09:09

do a pivot table in chat GPT, but

you need to decide when to do a tip

192

: 00:09:13

pivot table when it's appropriate.

193

: 00:09:15

Like when do I wanna aggregate

data based upon categorization?

194

: 00:09:19

And group buys, right?

195

: 00:09:20

That's something that you're still

going to need to do as a data analyst.

196

: 00:09:24

And that leads me into my

third prediction, which is that

197

: 00:09:26

your domain experience is more

important than ever in: 2026

198

: 00:09:31

And what I mean by that is like when

you look at a data analyst, they're

199

: 00:09:34

analyzing data, that's half of

their job, but then the type of data

200

: 00:09:38

is their other half of their job.

201

: 00:09:39

What's the data about?

202

: 00:09:41

Is it healthcare data?

203

: 00:09:42

Is it financial data?

204

: 00:09:43

Is it music data?

205

: 00:09:45

Is it marketing data?

206

: 00:09:46

Is it sales data?

207

: 00:09:47

Like there's always half of the

domain in a data analyst role.

208

: 00:09:51

And I think that's gonna matter

more than ever because once again,

209

: 00:09:54

the how to do your analysis is

becoming less and less important.

210

: 00:09:58

The actual skills, like the actual

analysis skills to, to do your analysis

211

: 00:10:04

are becoming less and less important.

212

: 00:10:06

What's more important is knowing what

to do, when to do it, and what the

213

: 00:10:10

results actually mean and, and how

to translate that to the business.

214

: 00:10:15

So if you've been a teacher before.

215

: 00:10:17

Like, you know how a classroom works.

216

: 00:10:19

You know how a school district works.

217

: 00:10:21

I've never worked in a classroom.

218

: 00:10:22

I might be better at data than you chat.

219

: 00:10:24

GPT might be better at

data analysis than you.

220

: 00:10:27

I don't think it is.

221

: 00:10:27

But let's, let's just for this argument's

sake say that it is, but it definitely

222

: 00:10:31

does not know your personal classroom,

your personal school district, or

223

: 00:10:35

our really, how a classroom or a

school district work in real life.

224

: 00:10:38

Like you've actually been in

the front lines and understand.

225

: 00:10:41

The industry, and that's gonna be

lly important for the rest of: 2026

226

: 00:10:46

And moving forward, you're gonna get

really deep and different, uh, data

227

: 00:10:51

niches or I guess industry niches,

and your knowledge is going to matter.

228

: 00:10:55

And I've, I've told the story before,

but when I worked for ExxonMobil,

229

: 00:10:58

um, at the time I didn't have

my master's in data analytics.

230

: 00:11:01

I had a bachelor's in chemical engineering

and I, there was these competitions,

231

: 00:11:06

they called them hackathons where they

would basically take everyone in the

232

: 00:11:09

company and say, Hey, here's a data set.

233

: 00:11:12

What can you do with it?

234

: 00:11:14

Like, what type of results

can you get for us?

235

: 00:11:16

What type of insights can you pull?

236

: 00:11:17

What type of tools can you make for us

that would be useful for our company?

237

: 00:11:21

And I'd enter these competitions.

238

: 00:11:23

And some people in these competitions

were literally like PhDs in computer

239

: 00:11:27

science, PhDs in mathematics.

240

: 00:11:30

These people were a lot smarter than

me in terms of computers, statistics.

241

: 00:11:35

Machine learning data, like these

people were really, really technically

242

: 00:11:39

and academically smart, but I was

able to actually win one of these

243

: 00:11:43

competitions because no matter how

much smarter they were from like a

244

: 00:11:47

computer algorithm, mathematics sense

than me, I knew the business and I

245

: 00:11:51

knew the domain better than them.

246

: 00:11:53

They had spent all this time studying.

247

: 00:11:55

They didn't know anything about chemistry.

248

: 00:11:57

They didn't know anything

about manufacturing.

249

: 00:11:58

They didn't know anything

about engineering.

250

: 00:12:00

And that's something that was my domain.

251

: 00:12:02

That's what I studied in school.

252

: 00:12:03

I had worked for the

company, like I understood.

253

: 00:12:05

I was like really hands-on with like

refining and manufacturing of gasoline

254

: 00:12:10

and jet fuel and stuff like that.

255

: 00:12:12

And I actually knew what was going on.

256

: 00:12:13

And so when I was analyzing the data,

I was able to analyze faster than

257

: 00:12:17

them because I actually knew, oh, like

this is what sulfur is, this is why

258

: 00:12:21

it's good, or this is where it's bad.

259

: 00:12:23

That would take them a long time

to actually figure that out.

260

: 00:12:26

Uh, and so I was able to work faster.

261

: 00:12:27

I was able to interpret my results faster,

and I was able to actually just come

262

: 00:12:31

up with better insights than they were

despite them being more talented than me.

263

: 00:12:35

And I think for all you career pivoters

who are listening, that's really exciting.

264

: 00:12:38

That's really refreshing

because your pivot actually

265

: 00:12:42

isn't a disadvantage in 2026.

266

: 00:12:44

It's an advantage.

267

: 00:12:45

It's what puts you above the

rest of the people around.

268

: 00:12:48

You like the stuff you studied

in school 20 years ago, the stuff

269

: 00:12:51

you've been working on the last seven

years that you, that you kind of

270

: 00:12:53

hate, you wanna get outta that job.

271

: 00:12:55

That information that you

learned isn't meaningless.

272

: 00:12:58

You can hold onto it and

actually becomes an asset to.

273

: 00:13:00

You

274

: 00:13:00

all this to say, I think 2026 is

gonna be a great year for you.

275

: 00:13:04

I think you have a great

opportunity to pivot in analytics

276

: 00:13:07

to level up in analytics.

277

: 00:13:08

I think people are kind of sleeping on

the analytics right now because let me

278

: 00:13:12

tell you, it is the bread and butter.

279

: 00:13:13

It is proven and there's so many companies

who are still under utilizing how

280

: 00:13:17

much they're doing data and analytics.

281

: 00:13:19

So while everyone's kind of

interested in AI and automation,

282

: 00:13:22

stay true to data analytics.

283

: 00:13:24

And you can use your previous

domain experience to pivot in and

284

: 00:13:28

use ai, but don't be afraid of it.

285

: 00:13:30

Like AI is going to be a tool that you're

going to be using down the road, but

286

: 00:13:33

it's not replacing you anytime soon.

287

: 00:13:35

Data analytics is far from over.

288

: 00:13:37

I think we're just getting started.