De-risk Your Content Strategy with MarketMuse AI

0:00

[MUSIC]

0:05

Hello everyone and welcome to Go to Market AI,

0:07

the future of your Go to Market tech stack.

0:09

I'm your host Sarah McConnell.

0:11

In these days, it seems like every company has AI.

0:13

But on this show, we want to go a level deeper so you can see firsthand how

0:18

businesses are actually applying AI to solve your business challenges.

0:21

We're going deep into the use cases and showing you

0:23

live demos of the latest and greatest in AI technology.

0:26

Today, I'm joined by Jeff Koyle, co-founder and chief strategy officer and

0:30

market muse.

0:31

Jeff, welcome to the show.

0:32

Thank you so much for joining.

0:34

>> Thanks so much.

0:35

I've been looking forward to this discussion and I'm very excited.

0:38

>> I'm very excited to see the demo, but before we jump into that,

0:41

I would love to know first,

0:43

who is market news, what do you guys do and who are you helping?

0:46

>> Absolutely. So market use is a content intelligence platform that sets

0:50

the standard for content quality.

0:52

So everything we do is about producing high quality content and

0:56

giving advice on how, what you should be creating, what you should be updating,

1:02

how to actually get into the details and show your expertise in your content.

1:07

So we give details about how to build content briefs, how to build outlines,

1:12

how to compete, but not just compete, but to differentiate.

1:16

And so we are for content strategists, content marketers, editorial teams and

1:21

editorial leaders, as well as search engine optimization professionals and

1:25

writers who are actually touching the content and ensuring that is going to win

1:29

We've been working with teams for now, gosh, almost 10 years to drive

1:36

improved performance in how much content that they produce succeeds.

1:41

While also improving speed to execute, we have teams that typically double and

1:45

triple the amount of content that they're able to produce, while also

1:50

improving their percentages for success.

1:52

So being able to improve those efficiencies while also improving volume.

1:57

>> That's amazing.

1:59

And as someone who owns our content function at Qualified,

2:01

I'm super excited to jump into this demo and see it in action.

2:04

So with that, let's jump in.

2:06

I'd love to see it.

2:07

>> Solid.

2:07

Okay, so, and I love Qualified by the way.

2:10

And I'm so I'd be absolutely excited to be able to review your site and

2:15

to build out some examples and content plans and recommendations.

2:18

Today, we're going to be looking at a demonstration of contentmarketinginst

2:23

itute.com.

2:24

And so everything we're looking at is through the lens of their site.

2:27

So one unique thing that market news provides is personalized information.

2:33

So we're not just showing you what the market says and

2:36

how hard something is generally for your to perform.

2:39

We're giving you insights on how your site, how powerful your site is, and

2:44

how difficult it will be for you to perform.

2:47

So that personalized insight allows us to be very predictive.

2:50

And we can say, if you focus on this, you just have to maybe write one or

2:54

two articles.

2:55

But if you want to branch into this new topic,

2:57

you're going to have to build an entire cluster or collection of content.

3:01

So we're going to start here in a situation where we've already made a

3:04

decision about what content we want or what topic we want to focus on.

3:09

And what makes market news really special is we're able to understand what it

3:12

means

3:13

to be an expert on a topic.

3:15

So we say, if you knew everything in the world,

3:18

if you were the most knowledgeable, number one expert on this particular topic,

3:23

what are the concepts that you would naturally cover in a page about that topic

3:29

What are the related concepts that you would have on your site as separate

3:33

pages,

3:33

as separate collections of content?

3:36

And we deliver that in an easy to read content, they're a topic list.

3:42

So in this case, we're looking at content brief templates.

3:44

And it's saying, if you were an expert on writing content brief templates,

3:48

you would also probably cover creative briefs.

3:51

You would also talk about search engine optimization.

3:54

You would also talk about various quality elements.

3:57

You would talk about content strategy.

3:59

You would talk about search engines.

4:00

You talk about freelance writers, because that's who you give a lot of these

4:03

things to,

4:03

creative assets.

4:05

These are things, internal links.

4:07

That's a good one, right?

4:08

These are things that you would give if you were an expert, you would include

4:12

them.

4:12

But if you weren't an expert and you actually never built a content brief

4:15

template,

4:16

you wouldn't necessarily know to include those things.

4:18

So your page might be very shallow.

4:22

It's just covering on keywords and basic keyword research.

4:25

So what MerkigMuse provides is really an expertise analysis and an experts view

4:31

on how

4:31

to cover a topic and how to know what intents are really important.

4:36

So in this case also, we can look at the ways that people understand this

4:41

concept.

4:42

So some people are looking for templates.

4:44

Some people want one specific template.

4:46

Some people wanted specifically for SEO.

4:48

They wanted in various formats.

4:50

They want creative.

4:51

They want video.

4:53

They want social media.

4:54

So these are the angles for my content plan.

4:59

So these are not focused on expertise, but they're focused on the user and the

5:03

searcher

5:03

and the way that they think.

5:05

So what we're also provide is who we're up against.

5:08

So the search results x-ray shows us the types of pages who we're competing

5:13

against in organic

5:14

search.

5:15

So we can say, what are the structures of those pages?

5:17

Do they attack this with images and video?

5:22

How long are these articles?

5:23

How many sections are they?

5:25

So I can get an understanding of what kind of narrative they're communicating.

5:29

And we also provide MerkigMuse's content score, which is our measure of quality

5:33

and comprehensiveness.

5:35

We take that a step further and we show a detailed breakdown of every one of

5:40

those competitive

5:41

pages.

5:42

So we can say exactly what topics do they cover on their articles.

5:47

So when you see red, it means that article is missing mentions of that topic.

5:52

When you see various colors, it means that you're mentioning it at various

5:56

levels.

5:56

So we can understand what are the things that everybody does and what are the

5:59

things

5:59

that nobody's doing that MerkigMuse says would differentiate us.

6:03

So I like to say in this view that there's money in the red lines because those

6:07

are things

6:07

that MerkigMuse says are very relevant and important to mention.

6:11

And they're also really important because they're ways that you can

6:13

differentiate your

6:14

page, which as a marketer, that's all we should care about, right?

6:19

Everybody knows what everybody's doing.

6:20

And this button shows us that.

6:22

This is basically the tail wagging the dog, the correlation view.

6:26

But MerkigMuse gives you the insights about things you might have missed.

6:29

I like to hear, I love it when people say, oh, this is my blind spot detector

6:33

with Con.

6:34

Yeah.

6:35

And it's hard to know those things without something like this telling you,

6:37

like I think

6:37

about any of our keyword research that we've done.

6:41

And a lot of times it's guessing.

6:42

Like we're, yes, your point of like, we're like, okay, we know we're talking

6:45

about this

6:45

topic.

6:46

And I think these are the relevant ones.

6:47

You can get some suggestions.

6:48

But to your point, it's not what the user searching for, nor does it tell you

6:51

what's missing

6:52

from those topics.

6:53

Like if we write about it and we know that keyword stuff, things not going to

6:55

do us any

6:56

good anymore.

6:57

That's not going to help us on Google.

6:58

So yeah, I, as someone who's gone through this project before, I can absolutely

7:02

see where

7:02

the benefit would be here.

7:04

Yeah.

7:05

And really when you're, um, when you're understanding the concept, then you're

7:09

understanding the

7:10

competitive landscape.

7:12

Now who do you want to understand?

7:13

You want to understand yourself.

7:14

Right?

7:15

What do you have?

7:16

So in this case, in this view, this is an 11 hour work process to do manually.

7:20

And we delivered in about 30 to 40 seconds.

7:23

This from left to right shows me my cluster.

7:26

So this is my cluster assessment.

7:28

So I know what my most relevant page is to this topic from left to right, how

7:33

big is

7:34

my cluster?

7:35

And am I really hitting it out of the park with any of these articles?

7:38

Are they all?

7:39

Are they really high quality?

7:40

In this case, a lot of red, right?

7:41

This is, we are not, we are, we do not have a cluster that's crushing it for

7:44

content-proof

7:45

templates.

7:46

We probably don't have an article targeting it.

7:48

We don't have a collective of articles.

7:50

So this is giving me insights that I'm all around this topic, but I don't have

7:53

anything

7:54

that's really nailing it.

7:55

So it's giving me insight on a cluster analysis.

7:58

It's giving me a gap analysis.

8:00

It's also giving me a point of reference of competitors, as well as what it

8:03

would mean

8:04

to be an expert.

8:06

So now I'm going to show you this from an editor's viewer or writer's view and

8:11

also

8:11

a page from marketmuse.com.

8:14

So we have an article that breaks down the most effective elements of content-b

8:19

rief templates,

8:21

shows you details, gives you actual examples, and Mark, in this case, we have

8:25

an article

8:26

that's performing extremely well.

8:28

So what Markmuse optimized shows us is the ways that we can take that

8:33

information, take

8:34

that information like we saw here and ensure that it is included in the article

8:38

that we

8:39

are building.

8:40

So we can get immediate feedback about the topics we haven't included compared

8:44

to the

8:44

recommendations.

8:45

We can also put ourselves head to head, I guess, any one of our competitors to

8:49

see

8:49

what they've done that maybe we haven't.

8:52

So we see absolute gaps.

8:53

We see gaps against our model of expertise and everything in between.

8:58

We also build our own custom artificial intelligence solution that gives common

9:05

workflows, like

9:06

improving writing, changing tone, building outlines.

9:11

We can even drive insights that can deliver competitive references and really

9:16

some custom

9:17

stuff that you can only do with Markmuse.

9:18

And what makes our artificial intelligence solution awesome is that it includes

9:24

mentions

9:25

and it keeps the context of everything about the research you've already done.

9:30

So it knows what it means to be an expert.

9:32

It has our topic modeling technology built in.

9:35

It has our knowledge graph generation technology built in.

9:39

So it's very special and it's providing insights from the lens of expertise.

9:42

That's amazing.

9:44

Yeah.

9:45

So then, we're looking at this now.

9:46

We're researching a topic.

9:48

We understand what's happening in competitive landscape.

9:50

We did a little bit of cluster analysis, but the job of a content strategist

9:54

and the job

9:55

of a content marketer is understanding everything that you have on your site.

9:59

And so we have an on demand content inventory offering called Markmuse

10:03

inventory and we

10:04

give you all the details about everything you've ever written, all the concepts

10:09

that

10:10

you've included and we provide unique action paths, unique differentiated

10:15

insights on how

10:16

to drive the most effective return on investment for content.

10:20

So what you're looking at here is we analyze content market institute.

10:23

We read 6,000 pages.

10:26

We extracted all the entities, all the keywords, all the topics out of every

10:29

one of those pages.

10:31

And that was 22,000 topics, which made up 29,000 page topic pairs.

10:37

And we allow you to slice and dice those, you know, that wealth of information.

10:42

We have sites with millions of page topic combinations that we analyze.

10:47

We're able to slice and dice that, build them into views, look at particular

10:51

topics and

10:51

collections of content and deliver some unique personalized insights on those

10:56

topics.

10:57

Most effectively in my favorite insight that we deliver is our topical

11:00

authority or topic

11:02

authority metric.

11:03

And that shows us a composite score for our breadth of coverage.

11:07

I like to say it's a cross and down.

11:09

Our breadth of coverage, our depth of coverage, the quality of our coverage.

11:14

That's the thing that most platforms forget.

11:16

Understanding how well we've covered something.

11:18

Have we done this with one page?

11:20

Do we have an effective amount of breadth?

11:22

Is it quality?

11:23

Does it have momentum?

11:25

Is it performing well?

11:26

And has the market stated that we are effectively implementing content?

11:31

So do we have social mentions?

11:33

Do we have off page factors?

11:34

We composite all that and that represents your business's competitive advantage

11:39

And we don't just do that like you've heard probably domain authority, right?

11:43

We're doing that at the topic page combination.

11:46

We're doing that at the topic level.

11:48

So you know, in this case, content marketing institute, they own what is

11:53

content marketing.

11:54

They have one of the most successful pages about that.

11:57

They have a huge band.

11:59

This competitive advantage is huge.

12:00

It'll be really easy for them to maintain or grow their presence on that topic

12:06

or related

12:06

topics.

12:07

We can also give them insights on the potential growth, if they were to focus

12:11

on this particular

12:12

topic and anything in between.

12:15

So we can also search this 29,000.

12:16

I'm just going to look for things related to briefs because I'm building brief

12:19

templates

12:20

today in case you were wondering.

12:21

I can see that they have 27 topics that relate to content briefs.

12:26

So in this case, this is a cluster which they could expand.

12:29

This is an article they wrote in 2019 and it hasn't been updated.

12:33

But they're still maintaining a lot of amazing rankings.

12:35

Imagine if you wrote an article in 2019 and it's still ranked number five in

12:40

organic search

12:40

for writing briefs.

12:41

Yeah, I want to protect that and I want to build a cluster around it so that I

12:45

can hold

12:46

on to these amazing, you know, valuable references.

12:50

So we can understand that really quickly and we can bounce to the most common

12:53

actions people

12:54

do when they're doing research.

12:56

So research topics, optimize this page, analyze my competitors.

13:01

We were just doing it, assessing our cluster, internal and external link

13:05

recommendations.

13:06

We can also look to prioritize semantically related topics right here while we

13:10

're doing

13:11

our research.

13:12

We can do this by looking at every single page that we have that's about this

13:17

topic or we

13:17

can just do traditional topic or keyword research.

13:20

In this view, I'm looking at buyer personas and this is an area where content

13:24

marketing

13:24

institute isn't exactly knocking it out of the park.

13:28

Their top performers are, you know, ranking on page two, 18th, 25th, they don't

13:33

have a

13:33

lot of presence here.

13:34

So market news can give them insights about things that they could win quicker

13:38

and things

13:39

that are going to require a larger content footprint.

13:42

So we take this from, I have no idea what to do.

13:45

I'm looking at my entire site.

13:47

We start to show you the insights and the recommendations about where you might

13:52

have an

13:52

opportunity or things that would be easier or things that would be bigger lifts

13:57

So I was just working on a site yesterday with a client and they were like, we

14:00

want

14:01

to own this topic and we need to own it by June.

14:04

And I'm like, well, you've got about 80 to 100 pages that you've got to work on

14:08

in that

14:08

timeframe in order to make any kind of an impact.

14:12

In this case, we're looking at buyer personas.

14:15

This is a situation where a small to medium cluster and some updates to some

14:19

pages for

14:19

content marketing Institute would drive immediate wins.

14:23

So we deliver that type of insights and that type of actionable next steps and

14:26

content

14:27

plans.

14:28

After you accept those content plans, you can start to build your content

14:32

briefs and

14:32

then when those content items are starting to be built and starting to be

14:36

drafted and

14:36

get immediate feedback from the research and optimized solutions that we showed

14:41

before.

14:41

This is only a small section of what Market Media does.

14:44

We also have solutions for Google News.

14:46

I did not show you our content brief solution.

14:49

Then we've got some amazing things coming down the road.

14:51

But this gives you a basic how a bill becomes a law.

14:54

What should I do?

14:55

How much work do I need to do?

14:57

Let's get going.

14:58

How do I make sure that I'm differentiated?

15:01

What content do I have that I need to make sure is integrated and considered?

15:05

And then now let's execute the page and ensure that it's going to win.

15:09

Yeah, this is amazing.

15:10

I really love the personalized difficulty to your point of like using AI, which

15:15

is like

15:15

I don't I always see difficulty.

15:17

I'm like, okay, well, like small, medium.

15:19

I don't know how, but the personalized difficulty compared to the competitive

15:22

advantage is such

15:23

a quick, easy way for me to digest.

15:25

Okay, I've got a pretty huge advantage here.

15:27

I have a big advantage that I can take.

15:29

I can use and it's not just like a generic difficulty.

15:33

It's for me how easy and like it's so easy to then cherry pick especially for

15:37

like strap

15:37

teams that don't have a lot of resources.

15:39

Like I can pick off the easy ones and I feel like such a good use case for AI

15:43

of just making

15:44

it so much more simple and time saving for you and making it more personalized.

15:47

Which I love.

15:48

Yeah, so I mean, just while you were saying that I created a quick view that

15:53

looks at

15:54

in this case, content marketing institutes quickest wins, right?

15:58

So we can actually look at something they're ranking on page two and we have

16:02

tremendous

16:03

amount of competitive advantage and low personalized difficult.

16:07

So these are quick updates or maybe one page specifically focused on this topic

16:12

and the

16:13

perfect intent and Mark amuses predicting up and to the right real quick.

16:17

So for a team that's strapped on resources, knowing what your quick wins are

16:21

that if they

16:21

were to go in and focus on visual storytelling and visual storytellers, it's

16:25

going to have

16:26

an immediate impact, strategic content.

16:29

They have a developing a strategy page that hasn't been updated in a while.

16:32

I know this.

16:33

It would be a quick win for them to get into building a content strategy

16:37

cluster.

16:37

So these having these things, imagine being able to go in.

16:40

I just had a discussion with somebody where we were able to find 15 examples

16:45

where they

16:46

had quick wins, executed on all 15, they went 100%.

16:51

So it's that type of effective from a predictions perspective.

16:54

This is very cool, Jeff.

16:56

Thank you so much for taking us through the demo.

16:58

If you're ready, I would love to move into our Q&A section.

17:02

For sure.

17:03

I love that.

17:04

Okay.

17:05

So first up is how long have you been building AI?

17:06

Oh gosh.

17:08

Well, we were in AI first company.

17:11

So it's been about 10, 11 years since we first kind of were into our inception.

17:18

I was a little bit of a late co-founder.

17:20

My original co-founders had built some basic infrastructure.

17:25

I came on to really bring it to market because I'm use case professional as

17:30

well as my background

17:31

in search engine optimization.

17:33

I've built a number of products.

17:35

That servers, search engine platforms, focus crawler stuff.

17:39

I've done way too many various things in the search world.

17:43

So we brought that to market, but it was an AI first company.

17:46

So our original, we have two patents on our technology and it's focused on

17:51

topic modeling,

17:52

graph analysis.

17:53

It's an ensemble approach to understanding what it means to be an expert and

17:57

that's

17:57

what we were founded on.

17:59

Now AI is, you know, there's 150 branches.

18:03

I think 30 main named branches of AI.

18:07

And you know, we were one of the first companies who was building in the

18:10

natural language processing

18:11

space.

18:12

We were the first business, I think, of our type to deliver a natural language

18:17

generation

18:18

or large language or a refined language model product for content marketing.

18:23

We actually pulled that product from the market and we're relaunching it in the

18:27

coming

18:28

months in a unique way.

18:30

But we've always been about two or three years ahead of the curve on every one

18:36

of our AI

18:36

innovations.

18:37

That's amazing.

18:38

I always love asking companies that are like AI first, how long are they been

18:41

building

18:41

and other product?

18:42

Because there was like, what?

18:43

The whole time.

18:44

I'd go, yeah, but we'll tell them.

18:46

You know, it's one of those funny things where my path to this was I had built

18:54

technology

18:55

in the past and had managed products.

18:58

I also led an internal search team for a large publisher.

19:03

And I take these user experiences and these behaviors and these workflows and

19:09

use cases

19:10

and think how can I make them amazing with artificial intelligence?

19:15

But also while putting the subject matter expert first in content is the key.

19:21

And in search engine optimization too, it's how do you take that person and

19:25

turn them

19:25

into kind of a superhero?

19:27

And everything we focus on now is making it such that the subject matter expert

19:34

who is

19:34

part of your team, the experts who are part of your team are going to be

19:39

invaluable.

19:40

And I think that's a really different approach and it's one that I really love

19:44

is we're not

19:44

automating because automation tends to yield lower quality content in our case,

19:52

right?

19:53

And when you're able to keep that human in the loop and focus on expertise, you

19:58

can automate

19:59

the things that are truly commodity and spend more time on the things that will

20:04

make your

20:04

business amazing.

20:06

So we free up teams so that they're able to spend their time and effort on

20:12

things that

20:13

cannot possibly be replicated.

20:15

Absolutely.

20:16

And is what you showed today in your demo, is all of that currently available

20:20

to customers?

20:21

Absolutely.

20:22

Everything you saw today is available and has been available inside our

20:27

platform for

20:28

quite some time.

20:29

We do have some things coming.

20:31

We're building an entirely new AI platform for content strategy.

20:37

And it does things that have never been done before in software.

20:41

I'm very excited about it.

20:43

Amongst those things, it automates some keyword research processes that are at

20:49

scales that

20:50

are impossible to even imagine.

20:52

So 150 to 200 hour workflows in minutes for understanding cluster analysis,

21:01

understanding

21:02

audience analysis.

21:04

We are building a kind of a platform for this is the precursor to fact checking

21:11

And that is a desperate need for a content teams where we are in the building.

21:16

We're going to launch it in its first phase coming up real soon.

21:21

Search results analysis as well as building content plans and content briefs in

21:27

a unique

21:28

way so that they're easy to digest for in-house or freelancers.

21:33

So those are all things that are coming.

21:35

But we have everything you saw and then five other things that you've never

21:39

seen before

21:39

that I showed.

21:40

So yes, if you have a content pain point from an ideation phase all the way

21:46

down execution

21:48

and want to get more predictive, if people in your organization think content

21:52

strategy

21:52

or content or SEO is a black box, we can give you predictive success.

21:58

And that's really a big game changer for most teams.

22:01

And now who are some of your current customers?

22:05

A couple I mentioned.

22:07

One of them I love is a company called Inside the Magic.

22:10

So Inside the Magic is a publisher for Disney related content.

22:17

And so they have a collection of sites.

22:20

They work with us with our news solution.

22:23

We have a news solution discover.

22:26

And that's actually something we're also going to be adding to is some news

22:30

related

22:30

insights.

22:31

And they are a huge success story.

22:33

If you go to Market Media's case studies you can find one.

22:36

They've gone up 50 or 60X as it's working with us as a publisher really changed

22:42

the

22:43

way that they do business.

22:44

We've also worked with a lot of B2B technology companies.

22:49

Agencies are a fastest growing segment.

22:53

In the past we were like we have no idea how to work with agencies.

22:56

And then over the last few years we've been able to really learn from them and

23:03

get to

23:03

the point where we can offer solutions where we're more flexible with how if

23:08

you imagine

23:09

once you just look at agencies they want to be able to look at a site maybe for

23:13

a proposal

23:15

or maybe they already have the client.

23:17

Maybe they're only doing one deliverable every year or quarter.

23:21

So that level of flexibility we offer for agencies and guess what?

23:24

We started to offer it for everybody.

23:26

So now you can look at your competitors.

23:27

You can look at your site one time.

23:30

Maybe you can't invest in having monthly content plans.

23:33

So we're very flexible with the way that we deliver the data and then in the

23:38

future the

23:39

way that we're going to deliver our documents and strategic insights.

23:44

Very cool.

23:45

And my last question for you is are there any other AI powered products that

23:48

your team

23:48

is currently using and loving?

23:51

Gosh, well everything I use is AI products.

23:56

And I don't know if there's a piece of software or technology that I use that

24:02

isn't.

24:03

You know, excluding some obvious things.

24:08

So using most of the prominent closed and open source APIs for language models

24:16

we built

24:17

well we referenced as a switchboard.

24:20

Some people call it a Delphi theory model or Delphi model.

24:25

We built that and we call it a switchboard where our technology can identify

24:30

the best

24:31

models to use for particular instruction sets.

24:34

And that's what makes us really, really special.

24:36

We are truly LLM agnostic.

24:39

We don't have to use one solution or another.

24:41

We also can use ensemble methods throughout any little use case, one you don't

24:46

even see

24:47

versus a big one, right?

24:50

We also use our own technology flip-down.

24:53

So everything we do at Market Muse is not just looking at what we built.

24:57

We're also considering are there other options to make sure we're delivering

25:01

the best quality

25:02

output.

25:04

We have some particulars about how we would select and things we won't do, et c

25:08

etera,

25:08

our internal policy.

25:10

But as far as like fun things that we use all the time, the embedding models in

25:15

OpenAI

25:16

are delightful.

25:18

Google Natural Language Understanding, the Google Knowledge Graph API are

25:23

magical.

25:24

OpenAI software, Matt Kudu is a great product for lead scoring, predictive lead

25:30

scoring

25:31

for a business.

25:32

I'm not just scratching backs here, but qualified is one of my favorite

25:38

software products ever

25:39

made.

25:41

It's totally is.

25:42

I think that it has taken the main pain points of many things and allowed you

25:52

to integrate

25:53

it with your CRM.

25:54

So if your CRM is optimized with artificial intelligence, it's naturally

25:58

prioritizing in

26:00

the same way as with those segments.

26:02

I love that.

26:04

But there isn't a piece of software that I use that isn't prominently

26:10

integrated with

26:11

artificial intelligence.

26:13

And so gosh, I could go on all day every.

26:19

Matt Kudu is actually going to be on the show here in a couple of weeks.

26:22

So it's not getting.

26:23

You're using them and then we'll also get to see a demo of them here in a

26:26

couple of

26:27

weeks if anyone's listening to this.

26:28

So that's awesome.

26:29

That's magic.

26:30

I'll tell you, there's a lot of tools that I use and they're part of my

26:33

everyday.

26:34

But the folks that I get most excited about are like some of these models or

26:40

these venture

26:41

studios or these agencies.

26:45

And one of my favorites that I, you know, everything they publish is called 923

26:51

and Andrew

26:52

Amman.

26:53

And because he's, he talks about some of his projects that he works on.

26:56

So like, like whenever I look at something, I want to build that.

26:59

I want to build that.

27:00

That's kind of like, I go and I'll give, I'll give one special one.

27:03

So there's a product called silly tavern.

27:05

Have you ever heard of silly tavern?

27:07

Okay.

27:08

So silly tavern is a product that was made to facilitate role playing games

27:15

with large

27:16

language models.

27:17

So it actually, you can build non player characters inside there.

27:23

But guess what?

27:24

You can use it for a lot of different things.

27:28

So I can actually define personas and put them in a virtual space and have

27:33

virtual focus

27:34

groups.

27:35

I can have conversations that go on and try to solve problems forever.

27:42

They will just come up with as many answers and questions.

27:45

So agent modeling, where we can actually create virtual space for agents to

27:50

interact is one

27:51

of my most exciting AI use cases.

27:53

And the fact that I can do that on my computer for free is just what it's, it's

27:57

quite shocking.

27:59

So I love this stuff as you can tell.

28:01

Well, that's amazing.

28:02

Jeff, thank you so much for joining us on the show.

28:05

I know you mentioned earlier that you showed just a few of the use cases.

28:08

So if you're listening to this and you want to see the other use cases,

28:11

obviously market

28:11

use.com, you can go check it out and get a full end to end demo.

28:16

Otherwise, Jeff, thank you so much for joining.

28:18

It was great to have you on the show today.

28:19

[MUSIC]