Table of Contents
- • No silver bullets: Building the analytics flywheel
- • Identity Crisis: Navigating the Modern Data Organization
- • Scaling Knowledge > Scaling Bodies: Why dbt Labs is making the bet on a data literate organization
- • Down with 'data science'
- • Refactor your hiring process: a framework
- • Beyond the Box: Stop relying on your Black co-worker to help you build a diverse team
- • To All The Data Managers We've Loved Before
- • From Diverse "Humans of Data" to Data Dream "Teams"
- • From 100 spreadsheets to 100 data analysts: the story of dbt at Slido
- • New Data Role on the Block: Revenue Analytics
- • Data Paradox of the Growth-Stage Startup
- • Share. Empower. Repeat. Come learn about how to become a Meetup Organizer!
- • Keynote: How big is this wave?
- • Analytics Engineering Everywhere: Why in the Next Five Years Every Organization Will Adopt Analytics Engineering
- • The Future of Analytics is Polyglot
- • The modern data experience
- • Don't hire a data engineer...yet
- • Keynote: The Metrics System
- • This is just the beginning
- • The Future of Data Analytics
- • Coalesce After Party with Catalog & Cocktails
- • The Operational Data Warehouse: Reverse ETL, CDPs, and the future of data activation
- • Built It Once & Build It Right: Prototyping for Data Teams
- • Inclusive Design and dbt
- • Analytics Engineering for storytellers
- • When to ask for help: Modern advice for working with consultants in data and analytics
- • Smaller Black Boxes: Towards Modular Data Products
- • Optimizing query run time with materialization schedules
- • How dbt Enables Systems Engineering in Analytics
- • Operationalizing Column-Name Contracts with dbtplyr
- • Building On Top of dbt: Managing External Dependencies
- • Data as Engineering
- • Automating Ambiguity: Managing dynamic source data using dbt macros
- • Building a metadata ecosystem with dbt
- • Modeling event data at scale
- • Introducing the activity schema: data modeling with a single table
- • dbt in a data mesh world
- • Sharing the knowledge - joining dbt and "the Business" using Tāngata
- • Eat the data you have: Tracking core events in a cookieless world
- • Getting Meta About Metadata: Building Trustworthy Data Products Backed by dbt
- • Batch to Streaming in One Easy Step
- • dbt 101: Stories from real-life data practitioners + a live look at dbt
- • The Modern Data Stack: How Fivetran Operationalizes Data Transformations
- • Implementing and scaling dbt Core without engineers
- • dbt Core v1.0 Reveal ✨
- • Data Analytics in a Snowflake world
- • Firebolt Deep Dive - Next generation performance with dbt
- • The Endpoints are the Beginning: Using the dbt Cloud API to build a culture of data awareness
- • dbt, Notebooks and the modern data experience
- • You don’t need another database: A conversation with Reynold Xin (Databricks) and Drew Banin (dbt Labs)
- • Git for the rest of us
- • How to build a mature dbt project from scratch
- • Tailoring dbt's incremental_strategy to Artsy's data needs
- • Observability within dbt
- • The Call is Coming from Inside the Warehouse: Surviving Schema Changes with Automation
- • So You Think You Can DAG: Supporting data scientists with dbt packages
- • How to Prepare Data for a Product Analytics Platform
- • dbt for Financial Services: How to boost returns on your SQL pipelines using dbt, Databricks, and Delta Lake
- • Stay Calm and Query on: Root Cause Analysis for Your Data Pipelines
- • Upskilling from an Insights Analyst to an Analytics Engineer
- • Building an Open Source Data Stack
- • Trials and Tribulations of Incremental Models
Sharing the knowledge - joining dbt and "the Business" using Tāngata
YAML is hard. We all want to involve our business customers more in what we do, but the command line is just too scary for many. In this session,
Chris shares the journey of the Trustpower team engaging subject matter experts across the business, and introduces Tāngata: an interactive dbt docs site.
Browse this talk’s Slack archives #
The day-of-talk conversation is archived here in dbt Community Slack.
Not a member of the dbt Community yet? You can join here to view the Coalesce chat archives.
Full transcript #
[00:00:00] Kelly Hotta: Welcome everyone. And thank you for joining us for the final day of Coalesce. My name is Kelly Hotta. I’m a sales director covering APAC for dbt Labs. I’m thrilled to be hosting this session with Chris Jenkins, sharing the knowledge - joining dbt and the business using Tāngata. Chris leads the data management team at Trustpower and talks about dbt a lot.
Let’s be honest. I think we can attribute almost all of the dbt fanfare in New Zealand to the combined efforts of Chris and Marcel. I met Chris a few years ago and when I was at Snowflake and his breadth of personal and professional interests really never failed to amaze me. He spends his spare time writing rap lyrics for corporate events, writing C plus enhanced YAML to automate his smart home and even ran for political office.
Before we start here’s how to make the most out of this session. All the chat is taking place [00:01:00] in the #coalesce-tangata chat channel of dbt Slack. Chris, could you have picked an easier word to pronounce. Anyway, if you’re not yet part of the Slack community, you still have time to join. Just head over to community.getdbt.com and sign up once you’re in.
Take a minute to set up the Slack chat next to your browser. Because this isn’t just for asking questions. You can add comments, you can share memes, emojis, and just have fun with it. When the presentation wraps up, Chris will be available in the channel to answer your questions. And Lee, our chat champion has already kicked us off with an icebreaker.
So make sure you enter your answer in that thread. Okay. Let’s get started. The stage is all yours, Chris.
[00:01:46] Chris Jenkins: Thank you. And thanks Kelly. Is it actually a shame that started because I was really enjoying the Slack conversation .If you’re not in there or really jump in. So I’m Chris, I’m a data guy [00:02:00] and really starts to be able to share here.
Let’s see, I have food here. So I am a dbt 0.14 . Class but today I want us to talk about how we joined the business together with awesome stuff that we make in dbt. Before I jump in a bit of an intro to me. I’ve hit always more, sorry, one moment. I can’t see my next slide.
I’ve had two major events in the last year that deeply impacted this slide deck and the presentation in general. The first was I got married in March. Now that’s a bit of a weird thing to try to intro a work related event with, but it makes sense in a moment. The other events that happened [00:03:00] this year, My LinkedIn connections passed my Facebook friends and number.
Now, these two are beams critically intertwined because the result of those events has been, you get to see my honeymoon for. So it is going to tie in throughout everything. Every photo here is mine. This is taken on the front streets of Glossier on the west coast of the south island of New Zealand. I don’t know if I mentioned that I’m a Kiwi.
I live in Toronto, so the east coast, about three hours south. So anything you see that’s in a honeymoon. It’s pretty hard to get here at the moment with closed borders. There was one way to get here. Come work with us. Just do it. Sorry, Tāngata . This is a hard [00:04:00] word to say. For context, this comes from today, the Maori language from the original people in New Zealand.
[00:04:12] Tāngata #
[00:04:12] Chris Jenkins: And this is something that over the last 20 years in particular, the younger generation have really started to be invited by the ability to come in and Hey, actually, this language is something that’s important. There’s meaning here that is really important to actually grow throughout the New Zealand culture.
And this word come is most commonly recognized in a particular phrase or a proverb, a Maori . And the proverb goes, if the heart of the harakiki was removed, the harakiki is as a flex plant. So at the heart of the harakiki was removed, where would the the bell bird sing. If you ask [00:05:00] me what is the most important thing in all the world, I would be compelled to reply.
It is people. It is people. It is people, hair, tongue tongue into, but that doesn’t quite cover it.
So Tāngata is not just about people. We’ve got this very simple Western concept of, Hey, it’s. you over there Tāngata actually talks about a much deeper concept. So we’re not describing mortality, it’s you and your generation, your generational group that comes before and comes after you.
So it’s not just. You and I, that Tāngata targets everyone that you do business with everyone that you provide for inside your organization. It’s your family at home. It’s your other groups that you’re all involved with? It’s your thing? So [00:06:00] Tāngata is a much, much deeper concept to them that. This is a one of my favorite places in times on the planet.
Not in Tato. We have a a famous beach called memo canoeing. And every year in April, we get up at about five in the morning and show up on this beach, around the cenotaph. And it’s this big color right here. And we stand together in different places, right across the country and right across New Zealand and Australia to remember ANZAC day. ANZAC day celebrates the people who came before us that gave out, gave their lives, what they believed in and this, the country that they believe should be a thing they wanted to keep New Zealand going in Australia, obviously. So when they showed up to a [00:07:00] war, they weren’t thinking of how do I protect this for myself?
that’s not how they’re guys, they showed up for theTāngata. They showed up for everyone that they cared about. That was what they were showing up there for. And see, I’m on this beautiful beach. This was about six in the morning. We’ve got 3000 people that showed up to remember what these people did.
We the descendants of these people that stood up and said, Hey, let’s we don’t want to do life the way that it’s going to be put on us. At least we do something about it. And this is the deep connection between something much more deep than, Hey, it’s people it’s actually that it’s that global concept that really, resonated with me.
The more I dived in. So this one word, the deeper it became. [00:08:00] So what’s this got to do with data.
So we are data people, we type things out of the different systems that run across their businesses and we make it useful when tune it into information that we can make decisions on. And these guys coming up on the screen, they’re actually coming for our jobs a little bit because. I’ve been at data a few years, not a long term, not the decades many that are here would have been, but back in the day, that stuff on the leaf to the paint used to be really hard.
We had to connect up and extract this data from Oracle and creates the data types and TDL for the place that you want it to go. Pull the, into the different lines of [00:09:00] transformation, whatever tool set you’re in, and we make sure all of these are running the right order to actually make the tire arrive properly.
And this doesn’t have to be done by people anymore. On the other side, it took a long time and Excel to do what you can do. One of these tools in a couple of minutes it took a really long time before that. On pen and paper, they just took actual years. It would’ve taken years to track some of the things that we can drag and, just get down to very, quickly.
You look back at the massive achievements that human kind has made the complexity of something like the addiction, the machine, and think I could probably sell that with my phone without any stress at all. And these things are just making our lives really, easy. I think putting us out of a bit of work.
So what do the people do [00:10:00] now? I’ve been really curious about that question for a few years now .In the future, computers should be able to figure out most of computers problems. So where is actual human value needed to be able to create something? And what we do, are we going to get to a future where we’ve got part of something, but we’re trying to do it the whole.
Speak about a time. when I first got this cool job job as a data and information manager. And I came from an analytics team side. I wanted to learn all of the history and understand the theory and the reasons why things are the way they are. And I read a lot at Kimball and I read a lot of Linstead and these sort of approaches and these some really, solid value in there.
I’ve never thought so much about the flight industry as well. I did while reading those books, it’s a great example, actually that they use with planes and [00:11:00] airports and things like that is something everyone understands. But for reading values, I started on an approach that it’s starting to get us somewhere, but it took a long time.
Because it’s trying to build a gold standard a gold standard outcome from day one. And in the dbt world, we actually get to iterate. We get to do something and then build it out some more and then build out some more. That’s what gets me really excited in the dbt space. I’m going off track, but. We turn up to the future and everything we do looks like the way we did it in the past, whether it’s the way that we designed our data models, whether it’s the way that we interact and design things, or whether it’s the way that we record information about our stuff that meets the data.
If we show up looking like this, then we might not get everything done that we might be capable of. Because all of our time has been freed up by [00:12:00] these tools that are just making life so much easier to get more done on the sides. We need to optimize the middle too. That is where humans are needed. Humans provide context.
[00:12:18] Context #
[00:12:18] Chris Jenkins: So being a data guy, and I’m sure everyone listening will probably relate to this idea. When people ask, what do you do? I’ve got a way of simplifying it because no I could probably fix you on us for you. Let’s be honest. I could probably fix your get rid of the virus on your computer from wherever you’ve been on the internet.
But that’s not what I do. If somebody asks me, I have to hit a simple way of explaining it. And the way I explained it is. My team takes data from all of the systems that know stuff about my company. And we [00:13:00] translate it for all those systems know and turn it into something that we can learn from as an organization.
So in the simplest way, we we’re translators, we’re providing context to things that the system has given us. And we’re translating from system language to, in my case, trust telling. And these are things like customers and services that we deeply care about inside of our organization. But the systems that we get it from don’t necessarily call it that.
They call it different things between them. And this is where the humans come in. I took this photo.
Bit of a weird photo to come across in New Zealand. I know some people won’t be as familiar with the country, but we’re not famous for our orangutan population. We’re a little too temperature for them [00:14:00] to enjoy themselves as much as they might in Borneo or something like that. What could we learn about this?
How do we place. And Chris’s honeymoon throughout the south island. We found some information. Here’s the meta digest from this photo that I’ve saved on. All of that tells us when it’s great, got the timestamps sorted. Name’s not going to help us much other than could food in the timestamps. I do see what phone I took it on.
There’s some details there. So if he wants to replicate it, you’ve got some of the details you’d need. And we’ve got a piece of information there that this was taken in. Hang the Springs in Canterbury. Now this is quite a famous place for the Hightouch. People coming from all over the world to visit painter, or is it going swimming in the many different mineral hot pools. [00:15:00] Hey, they had an option of the different Springs that the water comes from. One is more alkaline. One is more acidic and that doesn’t help us answer the question about the orientation. So we need some context from a human. Here’s that context. So this was taken at forest trust, tasteful protect styles on Google maps and windy tells us amazing sculptures.
So this tells us some, these are sculptures. It’s not a real orangutan. If you’d looked really closely, you probably knew that, but you still don’t know why it’s there. So we’ve taken the. The base information, we’ve got a photo of it or entertain. We’ve got the meta data. We did a coming from, and then we’ve got the human context that actually makes this useful.
And this has made it a story that [00:16:00] actually makes sense. It’s been translated from system into human language. Here’s a story without meetings, right? These photos are part of one story. I’ll give you a moment before I tied them together for you.
Photo on that lead was taken, and this is one of my favorite places. Actually, we drove from Dunedin downs on the east coast of south, island. We drove right down to the Southern most tip of the south island. This isn’t visited quite so often. We drove about eight hours a day. It was a very long drive and ended up at the end of that day in Queenstown.
So slight point is a place we not much groceries as you see in the photo. It’s the southernmost tip of what you’ve got is basically intact to go. If you keep on going south from here and the wind comes [00:17:00] out and the waves come up and it’s just the strength of the elements in this place is a night. So we drove way out of our way to visit this place that I just wanted to see from there.
We drove right up through the the middle of south Southland and into the Queensland. At the end of the day, we stayed at the Hilton. Fortunately. No international visitors are allowed in New Zealand. At the moment, I suddenly got a great group at a much lower rate than we would normally get. And they greeted us.
They knew it was our honeymoon so there’s this bottle of wine on the table. And I sat down to read the card that they’d given us and I say my shoe and I had walked into the Hilton hotel with sheep poo on my shoe, without noticing until I tracked it across the room. So I think it’s three photos to give a, tell a story that [00:18:00] this is one, picture.
This is I’ve given you the context. This is the story of my day and how, I felt in that moment as I was standing in the left-hand basin of this beautiful bathroom, three times the size of the room that I’m then washing the poo off my shoe in the basement. And that feeling is. Something that I could only tell through these pictures, one more piece of context with gaps, it’s like one to here, this sign points to where we are.
It tells us the acquaintance that way you’re facing the wrong way for me. South pole’s that way, it tells us where exactly we are. It tells us the meta data of this information. We are. 46 hours, 40 minutes, 40 seconds. I think that’s how that works. [00:19:00] Then I actually searched this, found out a little more about this particular location and those coordinates actually give the directions to about 250 meters offshore.
So just a brief note that even if you get great meta data, you probably should still check it. Now the other point here is there’s no point doing something over and over if it’s done before. So talking about context, there’s no point everyone’s doing the same thing. This is a tree in one account, which is visited by tens of thousands of people. And all of these people are repeating exactly the same thing is they will come on this tiny beach. At this very picturesque place. So in data, finding ways to capture something once and then replicate it to everyone, that’s the way that’s how we actually [00:20:00] generate value that lasts so that once someone takes the perfect photo of this trip, any titles that with comic saints in a haystack, this should be the only time that we need to document this piece of information.
And if somebody needs to change it, they come back and change it here. Maybe any piece of fault. So how do we do that with data? We need a place that we can record it. Keep on building on top of the shoulders of people that have come before us. The Tāngata that has come together to make this thing happen.
[00:20:43] Tāngata demo #
[00:20:43] Chris Jenkins: This is my favorite picture. Look at the story for this one. So, Tāngata and I have made a thing basically. So I spent my evenings early in this year. In about 200 hours I learned [00:21:00] node to write this. I learned react to write this and I’ve built a catalog that can do some cool stuff that I want to show.
This is essentially dbt docs. The intention of this is to not replace dbt docs. I’m not wanting to sell the store to get anything out of it, but I want to demonstrate what we do if we actually let people work together to take their perfect photo and share it rather than tens of thousands of people going back to the same place to try and find information.
Yeah. Yeah. We’ll scan. If you’d like to not begin and fall, if you give that to a business owner to say, Hey, can you document these sources that we got from somewhere? That’s going to be hard for them, but if we get them to come in here and click around a little bit.
And then that’s just straight away available in the [00:22:00] repository that you’re working in. Then you’re actually enabling people who might not engage with data anyway, to start to get engaged, to start to do stuff he out a catalog, I think should do with us. And what back to dbt. And I broke this actually to show up the people who aren’t doing that at the moment and because I like to prove things can be possible.
So we need to do stuff like that. You should just be able to click around Ames, not have to worry about whether it’s filmed the right thing, found the right file. Am I in the right place? If I doubled something up some winning and you should just be able to click around and let’s figure your way through and it should just generally make scenes like this.
That’s how you should be able to catch a meta data. It shouldn’t be scary. It shouldn’t look like this thing on the lift for someone in the [00:23:00] businesses describing a source, or even for a data analyst to know Louisiana, but wants to work at it faster. You should just be able to do this stuff. A couple of other things in here I wrote, but I think a call we’ve got some auto tech stuff here to do that.
It’s promoted models and stuff. We’ve got some deep emotion models. So when we search for customer, it’s going to put them into the bottom of the list and this is all available. If you PIP install, Tāngata and hip, it doesn’t support the McREL on top of the ice. So it’s just to gain GATF or jump into git hub Tāngata underscore local.
You can just jump in, have a look. It’s not great. It’s. That’s a little bit buggy. I haven’t done a unit tasty, but I really want to show actually guys, this is something that this is how we should interact with our stakeholders. We not all engineers, [00:24:00] not all data. People are engineers, not all data people can get.
And so it takes data with a black background and feel comfortable that the more as an industry and as a collective community inside dbt them, we get to stuff like this, that peanut placement. And that’s about all I have to say, except for where to from here steal my courage, build some stuff contribute.
[00:24:30] Where to from here? #
[00:24:30] Chris Jenkins: I haven’t got as much free time, so I’m not moving this heavily, but I do have time to build some intellectually start to, as a community say, Hey, let’s make this a thing. Lisa, our business stakeholders hit and miss with our stuff and we trust them. That’s the way we should be interacting with our businesses rather than owning it all in it.
And if you need to change that, we need a job. And if we can start taking a step that way towards the equation, [00:25:00] it’s going to be a much, nicer place to be in with the waves coming up in the window. The facts is this particular spot was one last piece of the context. See, if you can figure out the context of this first, and if you want a screenshot, here’s the dbt logo. I didn’t ask them about this. It’s been done.
[00:25:30] Q&A #
[00:25:30] Kelly Hotta: Amazing Chris. That was great. The chat is going off. I think you made Drew’s brain explode during your demo.
We have a couple of minutes, so I might ask a couple of questions if that’s okay. It looks like they’re mine. So I’ll ask one that I just put in the chat. And it’s one that I probably should.
But is this thing used in production at [00:26:00] Trustpower right now? And could you give us a sense of what the uptake was like were there any hurdles getting your less technical folks to jump in or was it pretty seamless? Yeah,
[00:26:10] Chris Jenkins: I’m not currently in production to Trustpower. There there are others using it some, and this is genuinely a proof of concept of how you here’s this thing uploaded the adage of the eighth, that it runs.
It doesn’t fall over. Oh, German teams to use it here, but I’m careful to keep my open source work as the things that I control sit very separately from the business. And just at the moment, that’s just how I’ve run this thing so far. I have used it here but I haven’t rolled with it.
[00:26:43] Kelly Hotta: Makes sense. I’m sure.
We have a few folks who would be good candidates for testing it out so we can chat. And then the other one that I threw in here was how do you determine the right level of [00:27:00] context to give your users without overwhelming them have had, do you have any tips or tricks that you’ve learned when approaching that problem at Trustpower.
[00:27:13] Chris Jenkins: So in reality, everyone’s got a different level of comfort. This design to this is probably to a bunch of people’s level of comfort. They’ve seen the tool that you can be surrounded with before. If you click in the field, you can change it just like Salesforce. We’ve used that before. It’s not scary.
So it really does depend you who you are, who you’re working with. Most of my customers internally are SQL writing analysts. And they not necessarily in code and get full on CI CD plotlines everything they do. And we’re moving in that direction. But the night they don’t know yet will fall as well. So for my customers, [00:28:00] that’s one group.
Another group is the people who know the sources really well and they go to databases, but they haven’t been linked again. So if we want to get a good documentation of sources, having that interface for them to jump in and actually use as the easiest way right now we see them a field list and getting a list of descriptions back.
It would be really easy, but much, much easier if we just gave them the mental effects of that.
[00:28:27] Kelly Hotta: Makes sense. Oh, I forgot one at the very beginning from Amy was are you going to be sharing any of your rap lyrics with us in the chat? You’re the one who gave me that fun fact.
[00:28:39] Chris Jenkins: Yes. So we don’t want to trust parallels in the tele sales team and someone from front line sales selling, fundamentally a service run up.
So now I’m running a data team, not quite sure how they happened. But in those early days, they leave things to [00:29:00] build, seeing the energy build morale, because you’re going to have people that say no to you. And it’s not very nice when people say no and doing so, having people alongside you, that you really liked that you’re energized by and.
Is great. There was a particular event that I ended up writing a rep for and a comic row into forming in front of about a thousand people at the end of the year organization to meet. I wrote one for my political campaign last year which. Yeah, that’s not on a chair here. Cause it’s very, slanted.
Obviously we do want to talk in context. But yeah, it’s just been a thing. I haven’t really got anything. That’s not slandered either towards trust are specifically all the politics side though. So I can’t really sorry.
[00:29:49] Kelly Hotta: No, that’s all right. I I wish I knew about this earlier, but maybe next time I call us we’ll yeah, we’ll definitely have.
Performed something for us. [00:30:00] Awesome. I we’re just at time. If you have any more questions, please enter them into the chat. It looks like there’s a couple in there, Chris. So if you can hang around and answer those, that’d be great. But thanks so much for sharing your story as always great to hear it and loved every moment.
And every photo made me very jealous and I can’t wait till we can get to New Zealand again.
Last modified on: Oct 11, 2022