0:00Another day, another platform
0:02today, it is Hex.
0:04Now
0:04Hex
0:06is
0:06a very interesting platform.
0:08It's not as commonly well known.
0:10In fact
0:11In most discussions, if you see Hex
0:13come up, um
0:16people
0:16people
0:17people probably learn about Hex for the first time when Hex comes up, if that makes sense.
0:21So for many of you watching, you maybe even never heard of Hex.
0:24Maybe you've heard it mentioned, maybe you've heard me mention it.
0:27Um, but it's a very interesting platforms, largely because of this uh concept that I'm gonna go back to again and again and again.
0:34which
0:35is its
0:35philosophical approach to the analytical workflow.
0:38This is the differentiator between all the platforms.
0:41And
0:42this is this is something that never gets discussed really in analytics.
0:45Everyone focuses on capabilities, but
0:48There's never this sort of broad brush discussion around the philosophic
0:51the philosophical standpoint of a platform
0:54and how that aligns with a company's culture.
0:57And so often
0:59you see people choose tools based on capability, but it massively misses because it doesn't philosophically align with the culture that actually operates in their company.
1:08whether that's how teams work or the way that people sort of um you know run their day-to-day job.
1:13I've seen this in virtually every industry.
1:15It's just day
1:16day in, day out.
1:17And for a long time
1:18BI tools uh
1:20were successful because there was nowhere else you could go.
1:22I
1:22actually think Tableau and Power BI
1:24have benefited from the very fact that they were standout.
1:27the best.
1:27So
1:28people generally decided to align with their philosophy, even though it wasn't necessarily the correct one.
1:34And this is sort of
1:35where
1:35we are today
1:36when suddenly every platform
1:38uh
1:38has
1:39uh you know
1:40uh
1:40improved
1:41the ceilings
1:41uh it
1:42like
1:43the ceiling's not really changed much, but the floor has risen.
1:46Um
1:46it's a lot easier now to find a platform that more closely aligns with a particular team's interests.
1:53or your whole organization's interests.
1:55And this sort of touches on another point, which is many teams have multiple BI platforms as well.
1:59So I say this preamble.
2:01Just to sort of set the scene
2:02because
2:03Hex
2:04in particular
2:05needs this explanation up front.
2:07Because what you're about to see is gonna make probably not as much sense to you out of the gate.
2:12But I I genuinely think
2:15in the direction
2:16and
2:18um
2:19time
2:19we're in
2:20with
2:21ai
2:22being this sort of tool that we use in our work, this sort of uh
2:25I call it
2:26this need to have some sort of analytical harness
2:30for AI tools.
2:32Hex
2:32is probably the best place
2:34one
2:34to take advantage of that moment to bring a cursor-like mindset
2:39to the analytical workflow
2:41But before we do that, it's
2:43it's worth just talking about what Hex is.
2:45Now I've actually had Hex on the channel.
2:47I had Rachel Herrera um on on
2:49the channel.
2:50She did a great session with me.
2:52Um, I actually really enjoyed this session the most because she's just she's just like me.
2:56She's in a Vadra.
2:57She's using Tableau.
2:59I
2:59you know, I'm always amazed when people like her say that they learned
3:02Tableau
3:03through me and then now they're doing great things at other places.
3:06Not that I have any hand in that, but it's
3:08It's inspiring to be able to know that the people actually try and help.
3:11I am actually helping and you know, they find their calling and they move on and they do incredible things.
3:16So I connected really well with Rachel when she was on uh yeah,
3:19another video with uh really great viewership, two point two thousand views five months ago.
3:23We went through the whole entire platform.
3:25I'll just unmute this so you don't hear it in the background.
3:27She covered pretty much everything, the whole analytical flow.
3:31So, what you're about to see.
3:33in this video today, I would urge you to go and watch this video if you want to see that sort of workflow more tightly managed
3:40But what Rachel showed back then
3:43was a hint as to where they were going.
3:45And obviously where they are now is completely different to sort of
3:49uh
3:50what she showed then and it
3:51I mean for every platform
3:52five months ago
3:53the world was completely different.
3:55So it's it's an inevitable sort of part of um
3:59uh the discussion
4:00but
4:01If I go back to
4:02the landing page,
4:04I've also been talking to Hex.
4:06They've not put me through a training like ThoughtSpot
4:09did.
4:09I don't think it's necessary.
4:10Hex is very different.
4:12And so you might ask her, well, still haven't really explained why you're interested in hex.
4:17It goes back to the way they work.
4:19The best way to think of it,
4:21you know, if I just sort of tear back Hex
4:24in terms of its philosophical approach.
4:26I'm going to simplify it massively by saying that it has a notebook mindset to the way that analytics is done.
4:33That essentially means that as you're building your analytics solutions, you're doing it in this sort of notebook style interface.
4:40And what that means is that your charts, the BI solution, the dashboard, whatever you build, runs as part of this package.
4:46And this package is uh think of it as like a folder or some sort of thing.
4:50And everything that runs off that that
4:53supports off that
4:54works off that package and so
4:57this mindset is actually really really valuable because I think there are teams in
5:01um the BI space.
5:03that are not you know
5:04serving you know
5:054,000 other people.
5:06They're not serving hundreds of other people.
5:09There are teams that are very small that are serving other small teams
5:14They need to be able to move quickly, they need to be able to sort of work quickly.
5:17And actually, if you look at the companies they support, it pretty much speaks to the point I'm making.
5:22Like
5:22some of some of the biggest names are in here.
5:25Notion, Anthropic, AWS, Cursor.
5:28A lot of startups that are working in very small, very capable teams, I'll call this out, you know, the kind of teams where one person is doing three or four or five different things that you'd go to a big company and find someone else to do
5:41They're using Hex
5:43a lot more.
5:45And
5:45this also leads to another sort of nice thing, which is it allows Hex to basically get away with doing a lot of really cool things
5:52all in one
5:53space.
5:54You can essentially bring
5:55um the kitchen sink
5:57to your analysis
5:58purely because they rely on one capability which is this notebook and if you can do it in a notebook then you can pretty much do anything you want.
6:04And that allows you to do some pretty crazy things.
6:07They just so happen to also have
6:09uh visualizations and BI solutions and stuff that comes on top of that.
6:13And of course, they're also big on apps.
6:14That's something that's
6:16pretty much common in every analytics platform.
6:18So that's my best attempt
6:20without really
6:22honestly
6:23transparently being an expert at this platform and trying to sell
6:27sell
6:27a an idea without
6:28really knowing much about the platform.
6:30Uh
6:30that's the kind of thing content creators do, I guess.
6:32Not not knowing much about anything, but you know, preaching
6:35preaching to you the audience.
6:37But
6:37it's my best attempt at sort of summarizing
6:39Why hex
6:40resonates in certain places
6:42and it also resonates with me in a very in a way I can't describe
6:46if that makes sense.
6:48I've always had this thing like, hey, if I could go on a desert island with one tool, it would be um
6:54Altrix, right?
6:55Because Ol Trix
6:55has been this tool that has, you know, just just helped me in virtually every way that I've possibly ever wanted
7:02That is no longer the case.
7:03And I've been trying to find like what what would be that BI
7:06tool that I go back to?
7:07And Tableau
7:08Prep is another one I often talk about from a data prep perspective where I'm always talking about, hey, I love the UX, I love the interface, I love the way it goes
7:15Hex
7:16is probably the one that I have this soft spot for where I just can't shake it.
7:20You know, people will tell me hey but hex isn't good in this scenario and this scenario and this scenario
7:25and I'll agree and then I'll still get back to it and I'll say
7:27but I still like it, you know what I mean?
7:30So
7:30I don't know what to make of that, but it has a hold on me.
7:34I can't tell you why.
7:35I'm still figuring it out and I do think it's very capable and I do think it's very good.
7:40So part of this video is to just walk you through the interface, walk you through how it works.
7:43I just have a trial environment, haven't got any special treatment
7:47like uh I had in yesterday's video with ThoughtSpot.
7:50Um, but what I'm gonna walk you through is just the sort of the vanilla experience.
7:54And actually I'm just gonna do the exact thing that you could do today if you
7:58went up here and hit get the get started and there is a free version that you can follow along with as well
8:03so you can uh absolutely go ahead and get started for free it's right here on the homepage
8:07before we do that I just want to sort of uh
8:10browse
8:11their
8:12um
8:13homepage because I think they do
8:15I'll also say this
8:16Hex
8:17do the best product marketing I've ever seen in my life
8:21full
8:21stop
8:22like there's no
8:24There is no doubt about it.
8:26If I could give an award to the best product marketing team in BI, it would go to hex.
8:31Hands down.
8:31No shadow of a doubt.
8:33I get what they're trying to communicate.
8:35They understand their users.
8:37They understand their products.
8:39There is no confusion whatsoever what space their tool fills.
8:42They're not afraid to tell you what their tool is not good at.
8:45They're happy to talk to you upfront about why they think their approach is good.
8:49They are happy to talk uh passionately about uh their stance on a specific area.
8:54And I saw this with talking to Rachel, but I've also seen it talking to other people in Hex
8:57They
8:58have some of the best uh creatives as well
9:00coming up with uh some of the best easy to use guides as well.
9:03Their documentation is really, really good.
9:05I love the way they work.
9:06And so
9:07Just looking at their website, if you honestly just actually engage with it and read it and work with it, you will get a sense of what their tool is about.
9:15It's the most straightforward product marketing I've ever come across.
9:19And honest to God, I think a lot of BI platforms could really learn a thing or two by uh picking up from it.
9:24Now they have this advantage because they know they sell to teams rather than to executives.
9:29So guess what?
9:29They get the freedom to be able to talk freely in that specific way.
9:33But needless to say, um, yeah, their
9:36their
9:36their
9:37their
9:37product marketing is absolutely fire.
9:39This is like a random tangent that I go off on
9:41because it's the kind of stuff I love.
9:43Now, you'll notice that a lot of their capabilities have sort of hinged around what I will say um
9:48uh
9:49agentic.
9:50sort of labels if that makes sense.
9:52If you went to all of these sites, even uh the other platforms I've covered so far and you just removed AI and Agentic, the the
9:58the solution would broadly stay the same
10:00So this is actually a really good summary of their platforms.
10:03You've got agentic
10:04notebooks.
10:04Previously it was just analytical notebooks.
10:07Conversational analytics, ask a question, get an answer.
10:10A context studio, build the semantic models, set up the things that you're going to use to allow people to do the work they need to do.
10:16Data
10:17apps, which is a fairly big thing for them.
10:19to be able to um build apps straight out of hex
10:22without having to do complex thing.
10:24The CLI is actually pretty cool.
10:26The CLI in a lot of instances today tends to be
10:29um
10:29a capability that you use so you can get clawed uh code to work on top of your platform.
10:35So
10:35CLIs are very, very useful because you're essentially creating
10:39A more efficient version of like an MCP if that makes sense.
10:42MCPs are quite token rich.
10:44A CLI just allows you to do the things that the platform can do straight out of the gate.
10:49And then they've got the products here on the left and they've got the capabilities.
10:52So they've actually done this really nice split of saying, hey, whatever language you talk, we speak your language.
10:57So if you're interested in our products, this is what you look at.
10:59If you're interested in what they can do, this is what that looks like
11:02like okay so you've got embedded analytics integrations
11:05business intelligence and exploratory analysis
11:07if you're gonna get hex
11:09I think you're coming to it for this capability the ability to go from
11:14I have an idea, I have a question
11:17to a chart, to something that you can share with someone that they can use, look at and make some sort of decision.
11:23And then on top of that, because you're doing it in this notebook sort of way
11:27Um
11:27you can apply what I would say
11:29CI CD principles to it fairly easily and get it productionized, even though you started out in
11:35in a scrapbook mentality.
11:37A lot of other BI platforms make you go through their own approval flows and ultimately
11:41because everything just gets approved, you never really know, for example, with a Tableau
11:45dashboard, like what is the best,
11:47like
11:48is this is this is this official
11:50or what is it?
11:51Um, just because of the way they're built.
11:52And so you get you have these things like labels where you certify them and so on and so forth.
11:56But
11:57anyway, you get my point.
11:58Um, because
12:00these things are being built by people who are very capable
12:03And the way they're built means that actually what's the what's sitting underneath them has to be good out of the gate.
12:10They tend to work um a lot better.
12:12In my opinion anyway.
12:13This is
12:13this is all assumptions
12:14that you don't you don't have to agree with me
12:16Um and again I would urge you to try the tools for yourself and and sort of get a hands on.
12:21So
12:22that
12:23is
12:23my
12:24thing.
12:24I
12:25with all the other platforms I haven't actually looked at pricing, but I will go here and look at pricing
12:29because
12:29There's a community edition and there's an individual plan.
12:32And this is actually the first analytics
12:34platform where they have this individual plan.
12:36And I look at this and I go, you know what?
12:38I'd actually use this as
12:39a data
12:40virus
12:41on my own.
12:41You know, if I'm entrusted enough to connect to data sources
12:44and just get going, this is an incredible offer.
12:4636
12:47Dollars per editor, you get the agent, five published apps, 30-day version history, median
12:52compute.
12:53That's that's pretty good.
12:54You can do what you need to, and then teams, their pricing is pretty comparable with other BI platforms
12:59Um
12:59but you get a lot more out of the box.
13:01Um it's actually
13:02genuinely you just get uh everything.
13:04There's no sort of secret tiers of enterprise and stuff like that
13:07So
13:07I think that's uh pretty good.
13:08And the resources are again really, really good.
13:11If we go to templates, this actually might be a really good way of just seeing some of the uh starter
13:16capabilities.
13:16Um
13:17they've
13:17got data modeling, uh
13:18data science, data visualization.
13:21Let's go to
13:23um
13:24I want to
13:25find something interesting.
13:27Let me see if I can find some
13:29um
13:31uh
13:31they're pre-built
13:33examples.
13:34In fact, no, let's not do that.
13:35Let's just let's just get into the guide.
13:36Let's just go.
13:37We've been
13:38going for 10 minutes and I haven't even got into
13:39the tool.
13:40Alright.
13:40So let me close some tabs so I don't get confused
13:44Okay, so
13:45oh
13:49so this is Hex, we've landed, and as is with most platforms now, they land you straight into their
13:54agentic
13:55um
13:56capability.
13:57So
13:57I'm going to select this
13:59sample dataset revenue drivers and we're just going to hit submit and you're going to see what it's essentially going to do out of the gate.
14:06So
14:06the first thing
14:08is
14:08is obviously going to
14:09it's going to do a couple of things behind the scenes.
14:11So what happens is
14:12it's running a notebook, okay?
14:13And so the notebook actually fire
14:15has to fire up compute
14:17Once it's then written SQL
14:19and it's got the necessary compute, it then essentially starts to run the queries and do the analysis.
14:23So that you know at a very basic level behind the scenes, that's sort of how it's thinking.
14:28So you'll see this that um
14:29how is revenue
14:30trending this quarter
14:31highlight any notable trends
14:33list the top uh factors that could be driving the trend and suggest how further
14:37how to further validate them so you get this sort of
14:40okay
14:40um if i click on this uh sequel six i think this is a specific connection
14:44sequel one
14:45sequel two
14:46uh
14:46dim
14:46customers uh dim
14:48accounts and dim
14:49opportunities so these are the different dimensional tables
14:52And it's doing a thinking kind of query here.
14:55So
14:55you can see that if I actually just ignore the response for a second and you you can see it's got the three cells.
15:01So these are the three that it
15:03picked
15:03And then it checked
15:04the current date and time to get a context.
15:08And it created
15:09this sort of analysis.
15:11And the the thing about this is if I just um
15:15open this.
15:16So
15:16This is a really good way of understanding how
15:20hex
15:21works
15:21because it's pull these three tables and then it's created what's known as a cell.
15:25Now a cell is like a
15:27uh
15:27like a type of insight if that makes sense so you've got this here
15:30sequel one
15:31select star
15:32from dim
15:33accounts
15:33let's go again
15:34sequel two
15:35select star
15:36from dim
15:36opportunities the kind of thing an analyst
15:38would do
15:39But then it's gone and created this calculation for us.
15:42So again, it's using this notebook mindset to literally do everything and then explain it to you.
15:48And it's
15:48it's showing you those calculations and then it's it's bringing in a visualization uh notebook in here and it's essentially drawing that up
15:57And
15:58this story on the left is a reimagination of this one on the right.
16:03So if you kind of match up these activities on the left with what's going on on the right, it's essentially what
16:09um you would be building.
16:10And before we had sort of this agendic push, Hex
16:13gave you this interface to be able to do this.
16:15So you kind of had to think in this very structured approach to how to you how you stack up.
16:20um your analysis.
16:21And to me it makes logical sense.
16:23I'll go back and say this is a philosophical viewpoint about how to do your analysis.
16:27And because it's done this way, again, if you're coming from something like Tableau, this might feel a little bit like
16:34because it's requiring you to think with uh
16:36what SQL queries am I going to run first?
16:38Whereas in Tableau
16:39you just connect and then you drag and drop and Tableau
16:41kind of does the magic behind you.
16:43And this is different.
16:44And again, it's philosophical.
16:46I'll just keep saying that because I think a lot of people will judge something like this very quickly out of the box and go, oh my God, Tim, you've lost me already.
16:53I was like, okay, yeah, but
16:55In some realms, especially in startups and small companies, this is how it works.
16:59This is how you do your analysis.
17:01So it is not too
17:03unfamiliar.
17:03But anyway, let's
17:04carry on.
17:05On the left, you'll see that it basically goes through, does all my analysis, and it brings uh various fields
17:10then.
17:11So let me uh go and do this.
17:13Let me sort of minimize that, and you'll see that it
17:15essentially goes through uh this and then we have a response.
17:19So
17:19It created this um
17:20if I click on that, as you can see it created this uh chart
17:24and
17:24the way that I'm um I'm sort of looking at this view, if that makes sense, is up here you'll see that you've got three tabs, agent, notebook and app
17:33builder.
17:33If I go back to the agent, you'll see that it kind of hides the notebook away.
17:37And if I go back to the top, it brings the notebook back in.
17:40So the notebook is the thing that's sort of driving my insight.
17:43So it's nice that you have this sort of
17:45um
17:47ability to switch um and I can close that sort of agent um
17:52panel and this would be sort of the interface that you're using.
17:55So you open up your analysis, you write your queries, you bring your data in, and once you've started to bring things in,
18:01You can then go to your app builder and your app builder essentially allows you to bring those specific things across.
18:07So you've got a generative app which is in beta, essentially allow AI to do it for you.
18:11Again, describe it.
18:12We saw this yesterday with ThoughtSpot.
18:14It was very sort of natural.
18:15experience
18:15but you can also go for a classic
18:17app
18:17for this specific uh
18:19video i'm gonna start with a classic app
18:21it's not exactly the experience they want you to do
18:23Um but you can go ahead and click oh
18:25this piece of analysis in here
18:27and um
18:28yeah
18:28okay it is giving you the ability to uh use a prompt but there's this uh
18:32all the problems don't want you to do this anymore.
18:34Please don't build, just don't build, please, trust me, Claude is
18:37better.
18:37But if you edit the app manually, you can see okay, we go into what is essentially the app builder experience.
18:42So um there's the agent, there's the notebook.
18:45which is this and then we can hide the agent and bring it back in.
18:48Then we've got the last thing which is the app builder.
18:50And now that I've actually instantiated the app builder without doing anything, I can hide the agent and this is essentially
18:56that view.
18:57Now I can actually go and work in this view.
19:00I don't have to constantly be um
19:02uh sitting inside of the other view so if I did want to build a dashboard uh
19:07from one of my existing um
19:09capabilities
19:10I can go in here, I can choose, hey, this is going to be my pie
19:13chart.
19:14And I can go and get the necessary things.
19:16So size, we're going to use, what we're going to use, what can I use?
19:20Count of records.
19:21Let's use that.
19:22And then horizontal split.
19:25We don't need a horizontal split actually.
19:27What I would like to do though
19:28is
19:29oh
19:29color
19:30by a region.
19:31That's what we'll do.
19:32So we needed to select those two attributes before the pie
19:35chart.
19:35Drew.
19:36So there we go.
19:36And then of course you can go into the style and change that and play around with it.
19:40So
19:41this sort of left
19:42hand, right hand side sort of panel thing, you've you've seen this pattern before in other tools.
19:46You just have to get your head around it.
19:48So
19:48hex, really, really simple.
19:50Notebook-based editor.
19:52You start with your notebook.
19:53Before we had AI, you started with your notebook.
19:56Once you built things in your notebook, you'd go to your app
19:59builder and bring things in from your notebook.
20:01You see it's a very sort of limited view.
20:03And this speaks to the point I was making earlier, which is if you
20:06used to doing data analysis in a notebook in Python in that kind of format then actually having your notebook right here allows you to do all that bit of analysis.
20:15and bring it in later on.
20:16And by the way, even once you're here, you still have this ability to um bring in things through the bottom here.
20:22So you can
20:22write your own SQL query and you'll see it'll just go ahead and create that.
20:26You can bring in a Python capability here.
20:28Again it will just go and create that
20:30text
20:30chart.
20:31pivot tables if you need to prep some data
20:34in order to get it to work.
20:36Parameters,
20:37so single values that you maybe enter and you can just play around with and then you can call those up.
20:42Then you've got data connections.
20:43Obviously, when you click on that, you get your standard data connector.
20:46And then
20:47you've got a bunch more different things.
20:49But needless to say, you've even got right
20:50back, which I'll call out because you know every platform calls that out now.
20:54But
20:54You get the mechanic, okay?
20:57So
20:57is this as easy as Tableau
20:59to pick up?
21:00Probably
21:00not.
21:01But is this powerful?
21:02Yes.
21:03because
21:03it gives you the raw capabilities in SQL and Python that you might want to use in order to do your work.
21:09Now the other thing
21:10this is, is
21:11it doesn't lock you in
21:13because ultimately you've still got the base code that you've written
21:17to do the work you need to do.
21:18So in a in a real world,
21:20let's say hex went away tomorrow, you'd still have some of this core capability you have in SQL queries that you've written and Python scripts that you're running.
21:29Hex is just plugging BI solutions on top of that code itself.
21:33So
21:33again, really, really kind of uh handy to have.
21:36So
21:37That is a whistle stop
21:39tour of
21:40I guess the mechanic of Hex.
21:42Um it needed a bit of upfront explanation, but
21:45hopefully
21:45you also understand this sort of um
21:48idea
21:48this journey so you start with a um
21:51uh
21:51a notebook
21:52in that notebook you have multiple cells
21:54each cell is a type of analysis
21:56when you've combined enough cells and you've done your data analysis, you've done your drilling and you've built it all out, you then take that to the app builder.
22:04The app builder then gives you the ability to bring those specific
22:07stories
22:07in and by the way this is just like a standard um
22:10you know
22:11um
22:12this is just a standard sort of a dashboard experience so you can
22:15you can go ahead and um
22:17Drag these around, swap them around.
22:20Um
22:20also
22:21you can uh
22:22change the way this works.
22:24So hold on
22:25if I go back and I do
22:27this
22:28um
22:30Okay, what I've done is I've cleared the dashboard.
22:32If we go back to the notebook and we go back to the chart
22:36just to show you how this
22:38actually
22:38works.
22:39If you go to the chart that you want to bring in, obviously when you're here in the cell
22:42view you have all the sort of raw elements that go towards this
22:45but um you can also toggle between a table and the chart which is uh kind of handy but actually if you click on this
22:51small icon you can add it to the app builder and it goes there
22:54and then um
22:55I think I believe uh I built a pie
22:57chart so you click on that it goes there go back to the app builder and there they are
23:00so you get this um
23:02ability to be able to uh
23:04you know
23:05Bring and drag and drop different things.
23:07You can also add tabs, so maybe you have different pages in your analysis.
23:11This is pretty straightforward.
23:13And um
23:13like Sigma, they actually have a pretty good lineage view.
23:15So if you go
23:16go up here to the top
23:17and um
23:18we just move this to the left
23:20hand side
23:21we can actually see
23:23the queries
23:24that were run so we
23:26some of the analysis
23:28prior was just straight off the base table.
23:30So sequel one, we didn't really do anything from that.
23:33SQL six, we just basically went and got a a table and it's basically numbering the sequel.
23:38uh
23:38queries
23:39as we did them.
23:40Um but this one was actually an interesting one.
23:42So we did two uh queries and then we actually took one of them and turned it into a chart.
23:47So
23:47it kind of explains nicely visually
23:49what that sort of uh
23:51analysis flow looks like.
23:53And I think it's um I think it's an interesting kind of framework.
23:56So
23:57Can I be honest with myself and say, do I work like this today?
24:00No, um, because I've largely been working in platforms like Tableau and Power BI.
24:06Which I'll say have like a WYSIWYG-like mindset to them, which is they they behave like a CMS to this kind of stuff.
24:12You see, Tableau's never really wanted to ever show you the SQL query.
24:16it's been creating the interface for you to do the work you do.
24:19But again, um
24:20if you've um you know
24:22been on YouTube, you've
24:23worked in like uh a a
24:25an
24:25analytics
24:26company that requires you to know Python and SQL
24:29like very, very well in order to do work.
24:31Um startups tend to have this requirement.
24:33Hex
24:34makes a ton of sense.
24:35It just makes a ton of sense.
24:36Okay.
24:37So
24:37I've got myself lost in here.
24:39What I'm gonna do is I'm gonna go open the side menu and now you'll see sort of
24:43the different uh spaces they have.
24:45So let's go back to the home page
24:47and this is actually where you land if that makes sense.
24:50So
24:50you will land in lots of different places.
24:52You could see I was actually back here in April
24:552026.
24:56This is my first project.
24:57So I think I had a couple of other
24:59demo um
25:00dashboards.
25:00Let's just open one of these up.
25:02Oh, look at that.
25:03I was building a nice little uh dashboard there to
25:06to do stuff and uh you know
25:07these things can look good um
25:09they don't like
25:09just because you're in a tool like this doesn't mean um you can't get them styled and formatted exactly how you want
25:16That is absolutely
25:18part of the course.
25:19So
25:19there's a demo one.
25:20What's another one I
25:21did here?
25:22This must have been like a
25:23oh no, these are
25:24I never built these
25:25These are applications slash um
25:28pieces of analysis that were done for me, I think, when I created the
25:32um
25:33uh
25:33the
25:34uh environment.
25:35So
25:36Now another thing about these apps and this sort of interfaces, when you go to share them, for example, you can absolutely share them to someone, you can embed them, you can add the sharing controls, you can share them to a group.
25:47That all makes a ton of sense and they become then things that people can browse and engage with if that makes sense.
25:53So um
25:54this is all to say.
25:55That's essentially the sort of workflow.
25:57Now you've got a couple of other things
25:59got
25:59ways to organize
26:01um
26:01various um
26:02aspects.
26:03If I just go back to my home page, it kind of gives you the
26:06uh let's say
26:07the lowdown
26:08if that makes sense.
26:09So here your projects, these are individual things that you've kicked off.
26:13Threads I guess are just discussions, data, connections.
26:16I don't really have any because I haven't set anything
26:18up.
26:18I haven't set up any uh favorites and then uh
26:20popular this week.
26:21These are just me kind of building and playing around with different capabilities.
26:25Now
26:26if I go back to um this side menu, let me just go to collections
26:31I don't have collections in this environment.
26:33I believe this is an enterprise setting, but essentially
26:35allows you to bring together assets from different places or different
26:38different projects if that makes sense.
26:40So think of it as a way of sort of clumping together bits of analysis to form a bigger
26:44bigger story.
26:45Now the final place that I I want to touch on, I don't believe I actually have this in my license
26:50is the context studio.
26:52But that is, I'm gonna say in a very sort of simple nutshell, that is hex.
26:56There is not much more to it.
26:59other
26:59than obviously the abilities that they're bringing forward into the platform to be able to build things like apps, um, build uh really complex types of analysis
27:09and it's pretty robust.
27:10There's
27:11with
27:11all these platforms, whether it's uh Sigma, ThoughtSpot, uh
27:15Hex, um
27:16Omni, uh
27:17Golden, when I get to look at that, um, they all let you do the same thing.
27:21They all have the same sort of feedback loop uh
27:25setup.
27:26in
27:26their environment.
27:28The reason I like Hex
27:30is because I think
27:32I can start to build my own sort of workflow and it doesn't punish me for the way that I work
27:39Even with Thoughtsbot
27:40yesterday, if I just sort of go back to this, you know, I kept on talking about wanting to drag and drop, but that wasn't sort of an option, right?
27:48In Tableau, um, you have these sort of sometimes really frustrating loops where you have to create a connection, then you have to publish it, and then you
27:56You have to, you know, you have to go through certain things every single time.
27:59And yes, you can centrally create them, but someone always has to do that.
28:03Um, you know, in in Sigma they solve that differently.
28:06You create a connection centrally, you give it uh
28:08access to a group of people
28:09then
28:09they come in and choose what they need from that connection and then go and build it up.
28:13So they're sort of broken down.
28:14Thoughts
28:14uh Thoughtspot
28:15does the same thing.
28:16So all of these philosophical differences, um, I think I think they mean more than they
28:22they've ever
28:23meant.
28:24um
28:24in today's world where AI is just massively making all of these things really easy to build.
28:29So
28:30I think I have uh gone as far as I can go.
28:34I need to get my environment correctly set up.
28:36I need to add connections.
28:37I need to do so much more to get this to a place where it's actually worth doing a demo.
28:42You can see here I've actually got nothing.
28:44in my demos.
28:45I've got I've got the standard um demo data that they give you, but I haven't got anything more than that.
28:50So we're gonna go back to the homepage.
28:52I'm gonna leave you here, I think.
28:54It's actually been a shorter video than I wanted it to be.
28:56I did want to go in and do more, but I'm realizing, I
29:00know, I need to
29:01I need to do a lot more than I did with some of the other platforms
29:04uh just to get like a decent environment set up.
29:08I don't think any
29:10of these um demos are also really representative examples of what the platform can do
29:16And
29:17if I just open
29:18something that I
29:19sort of messed around with a while back, yeah, they're also
29:22unless you're sort of part of this um
29:24process, it's really hard for you to kind of
29:27um
29:28see
29:29like
29:30w
29:30look
29:31what went behind making this happen if that makes sense, right?
29:34So
29:34um
29:35even if I go back to the edit experience here, you go back to the same loops and I think I've already talked through this in enough detail.
29:41So um
29:42yeah I think I'm gonna stop here.
29:43I'm gonna stop here.
29:46I am gonna spend a lot more time, I think
29:48having to set up hex, probably more than I did other platforms.
29:52And by that I mean set up a data model, set up um
29:56some
29:56connections
29:57and then uh
29:58build out
29:59an example and the reason it will be harder with hex
30:02is because i'm not building out example that in the way i'd build it out with tableau
30:06I'm gonna build it out
30:08trying to do the thing that they want you to do, right?
30:10And so that requires a completely different skill set.
30:13Like I've never written SQL on screen before.
30:15I should never have to do that
30:17Um but I think it
30:18and y
30:19by the way you don't have to do that.
30:21You can use
30:22um
30:22their interface to sort of bring those those things in.
30:25But even just the
30:26the process of thinking through those steps.
30:28I wanna set that up in a fair way to the product.
30:31that makes sense and Rachel has already done a video on the channel which is why I called that out at the start
30:36go to that to kind of see that workflow working in earnest
30:39I think it's a much much better place
30:41but
30:41I'm gonna leave it at this.
30:43I still like hacks.
30:44It's so
30:45it's so unusual for me to like something this.
30:48I'm gonna say a little bit tricky.
30:49It's a little bit tricky.
30:50But if you have the right skill set, you're very comfortable with SQL, very comfortable with Python, you're working in a small team, you love this sort of notebook mentality, and you want to just be able to build things and ship them really, really quickly and just have a BI
31:04tool
31:04sit on top of that.
31:05I think this is a really, really good work for it.
31:07And in the agentic world we're moving to where
31:10can
31:10Claude can do some of this for you?
31:13it could be a really
31:14really
31:14really
31:15powerful harness
31:17for some of that ai
31:18goodness that we want to bring into our tools and you saw that earlier on when I went in to try and do the app builder
31:24It asks me, hey, what model
31:26do you want to use?
31:27That's really the experience that I think Hex
31:29is heading to.
31:30And that's going to be a pretty wild, wild place.
31:33Anyway, thanks for watching.
31:34I'll see you in the next one.
Hex | Tool strategy | Data visualisation | AI & ML | Analytics 
