4#15 - Säde Haveri - The Data Governance Framework (Eng)

«I consider this Data Governance as a cure. (…) Data Governance can make things better.»

In this clarifying conversation, Finnish data expert Säde Haveri shares her 18 years of experience and introduces a practical framework consisting of five key elements that can guide any organization's data governance journey.

Säde, who is a Data Governance Manager at Relax Solutions and co-founder of Helsinki Data Week, first explains the important difference between a framework and a playbook. While many consultants offer ready-made solutions, Säde argues that a truly effective framework functions more like scaffolding, helping organizations uncover their own best path forward.

We dive deep into the five elements: the choice between a top-down or bottom-up approach, the balance between defensive and offensive strategies, how to define the right scope, identifying key stakeholders, and the strategic role of external consultants. Säde illustrates how these decisions affect the structure, implementation, and success of data governance, with practical examples from his own experience.

Here are our hosts key takeaways:

• Data Governance is at the heart of the socio-technical system - it requires a variety of skills.
• The experience for the end user has not change much in the last almost 20 years.
• There is a need for «group support» for data people.

What is a framework?

• There are ambiguous connotations of the word «framework».
• A framework is not a playbook.
• A framework describes the what, not the how. You need two adjust it to your reality.
• Think of a framework as non-prescriptive.
• Frameworks are related to best practices, but they are not the same thing.
• Use it to identify your strengths and build your data governance practices around those.

Top-down or bottom-up

• Can be a management awakening (e.g. GDPR), or a need for better data at a practitioner level that initiates the need for data governance.
• Top-down often materializes in conceptual approaches.
• You start at a conceptual level, you will create data governance roles around these conceptual entities.
• From a bottom-up perspective you are building governance around your tables or datasets.
• As a middle way, you can focus on data products as objects to build governance around.

Aligning strategy defensively or offensively 

• Defensive vs. Offensive strategy - based on an article from Davenport 2017.
• There is no one-size fits all.
• You need to understand your motivation? Is it build due to risk mitigation needs or for business value creation?
• You need to understand your sector-driven differences.
• Look at this as a spectrum, where your approach can differ between offensive and defensive based on the criticality of the dat for the use case you are working with.
• You always have to show your value, the value needs to be measurable.
• AI ready data can be both offensive and defensive.

Identifying Scope & key stakeholders

• Identify your stakeholders and scope based on the strategic alignment on offensive vs. defensive.
• Use business stakeholders rather than IT, to gain a better understanding of the underlying problem.
• Data Governance is rather a business value enabler than a cost saving activity.

Determining the role of external consultants.

• You have to sell your solution. Selling is something you cannot outsource.
• If you are looking at tooling, consider if you can find consultants with the right knowledge and capabilities.
• Try to understand what experiences the consultants can bring to the table.
• Ensure that you are aligned on a methodological basis.

Transcript

Introduksjon til Metadema-podcasten

Speaker 1 0:00

Dette er Metadema, en holistisk syn på datamanasjon i Nordisk. Velkommen, jeg heter Winfried og takk for at du har lyttet til meg i dette episodeet av Metadema. Vår visjon er å promisse datamanasning som en profesjon i Nordisk. Vær så god du er, og det er grunnen til at jeg inviterer nordiske eksperter i dat og informasjonsmanagning til å snakke. Velkommen til Metadema, og sesong 4 er en veldig interessant fordi vi begynner med noe som er en pattern her. Jeg vet ikke om det er av grunn av det er ikke planen fra min side, men vi har Danmark, finland, danmark, finland på podcasten, hvilket er litt interessant. Så jeg har Sabi med meg i dag. Jeg håper jeg har pronunert deg korrekt. Hun er fra Finland. Hun er en del av det vi kaller Helsinki Data Mafia. Jo Ries nevnte det finländske data-kommunitetet etter Helsinki Data Week i oktober. Sadi har jobbet med datagoverning og hun har jobbet med datagoverning for et langt tid. Så vi skal snakke om et datagoverningsspråk, et språk som Sadi bruker i hennes arbeid, som hun synes er framework The framework that Sadie uses in her work that she finds effective.

5 elements of Data Governance

Intro Säde

Speaker 1 1:25

It's about five elements of data governance, and let me quickly run through those five, but then I will give the word quite quickly over to Sadie after that. So the first element is top-down versus bottom-up. The second element is defensive versus offensive, and these two elements are the ones that we are going to talk about, i think the most in this podcast session, but there are more. The third is find the right scope, the fourth is key stakeholders. And the last one, the fifth, equally important, is what is the role of externals? So, before we dive into the topic. Og det femte, som er like viktig, er hva som er rollen av eksternal? Så, før vi går inn i temaet, så tror jeg det er mye å utforske om Finland, om datakommunikasjonen i Finland og, mest av alt, om deg, sadi. Så glede deg til å introdusere deg selv.

Speaker 2 2:16

Hei og det er fint å være her i dag. Så mitt navn er Sadi Haveri og My name is Sede Haveri, and I have been in total, working with data for 18 years now in different kind of fields. I currently work for a company called Relic Solutions which does supply chain optimization using machine learning models. I work there as a data governance manager. I also do some freelancing on the side and I'm one of the founders of Helsinki Data Week, som du nærmest nevnte i begynnelsen. Jeg er også en børnemmer av DEMA Finland, så dette er noe som vi har, som vi sammenligner, eller DEMA sammenligner oss. Jeg tror at med datagoverning har jeg jobbet i fem år nå. I think that's with data governance. I've been working now for five years, i think. I started as a data warehouse and BI developer, so I actually come from the tech side, which is not that common in the governance sector.

Speaker 1 3:18

Very true, and I think this is what makes the governance so interesting. I think that there is such a mixture and also such a need of mixture of people, of competencies, of backgrounds in data governance. I described it as data governance very much at the heart of what we call a socio-technical system, because you have both the social side and the technical side very much combining and finding their way together in data governance. For me, this is what makes it so interesting. I don't know what do you think, både den sosiale og tekniske siden mye kombinerer og finner seg sammen i datagovnen. For meg er dette det som gjør det så interessant. Jeg vet ikke hva du synes.

Speaker 2 3:50

For meg hvordan jeg faktisk driftet inn i datagovnen er at, som jeg sa, 18 år med data Og hvis jeg tenker til da jeg begynte 18 år siden som junior, hvis vi tenker på end-usere av data, så var det problemer med hvor jeg finner data og hvem som kan fortelle meg mer om dette dataet og er dette dataet tilgjengelig for min brukerskap? har det alle rådene eller har det bare en liten skåp som noen produkter eller noen steder, eller hvordan det fungerer. Og hvis jeg går tilbake, eller tilbake til 18 år, så tenker jeg på end-user. De spør de samme spørsmålene Hvor er min data?

Speaker 2 4:40

Kan jeg bruke den for dette brukerskapet? Hvem kan fortelle meg mer om det? Så teknologisk, hvordan vi arbeider med data og pipeline. Vi har cloud, vi har BI-solutions, vi har selvservice alle disse tingene og utviklingen har endret seg mye, men opplevelsen av end-user har ikke endret seg så mye, og dette ledde meg til datakataloger. Jeg ble veldig enten om dette, fordi da jeg så teknologien og da, kanskje fem år siden, begynte det å utvikle mer, også i Nordiske, og jeg tenkte at ja, siste noe for end-user, dette er greit, men da ok, etter å ha gjort noen projekter er det ikke så lett å få dataloget i den formen at det faktisk a few projects.

Speaker 2 5:20

It's not that easy to get the catalog into the shape that it would actually produce the value that I see that it has the potential to do, but the implementations are maybe not there yet, but very quickly. When working with this I realized that if you don't have the data governance, then the catalog thing isn't going to work either, because you need the people to kind of who own the metadata and the understanding of the things. So this is kind of like my, for man må ha medlemmer som kjøper metadata og forståelse av tingene. Så dette er min vei, eller hvorfor jeg har drifta i denne direksjonen fordi jeg ser potensialet her for noen kjører for gamle smerter.

Speaker 1 6:00

Det er en fantastisk ansvar. Er det også grunnen til at du ble en del av DEMA Finland?

Speaker 2 6:08

DEMA Finland er at. jeg faktisk gjorde kjøretjenesten CDMB først. Jeg gikk ikke til DEMA immens Jeg vet ikke om det var en DEMA Finland Jeg gikk til den internasjonale bransjen. I didn't actually know that there was a day in Finland. I joined the international branch but I found out later that there is a day in Finland and met with some of the people and after that I joined. because if there is one thing that I very often talk with like the data governance people in Finland is that there is a lot of need for group support For like having other people that you can talk to, that know what you're going through, because usually there is one data governance manager, one data governance people or two in a company. Everyone else the role holders are like from inside the company. So I think this collective support is what drove me. holdere er fra innenfor sammenhengen. Så jeg tror at denne kollektive støtte er det som drev meg til DEMA, eller nødvendigvis Det er litt som gruppeterapi.

Speaker 2 7:15

Ja, exakt, gruppeterapi. Jeg var søkende for Takk, du fikk et godt ord.

Speaker 1 7:21

Hva gjør du når du ikke arbeider med data og datagoverning? Hva er good word? What are you doing when you're not working with data and data governance? What are your hobbies?

Speaker 2 7:29

I read a lot, or audiobooks mostly. I also attend quite a lot of data seminars but I don't know if you can call that a hobby. The Helsinki Data Week? I do a lot of that. I don't know if you call a company a hobby as well, but I like meeting people. I do a lot of that. I don't know if you call a company a hobby as well, but I like meeting people. I like connecting people, i like introducing people to each other and I like learning new things. So the books, and also like biking and walking and doing anything with my son.

Speaker 1 8:03

Fantastic, and now you part the way already into the main topic, but I think, before we can talk about your five elements of data governance, i think there's something important we need to settle, and this is a question that comes up quite a lot What is a data governance framework and why do we need it?

What is a framework?

Speaker 2 8:23

I really love that you asked this question. I think it's so great because especially the word framework. And why do we need it? I really love that you asked this question. I think it's so great because, especially the word framework.

Speaker 2 8:28

I have to admit that I've been one of those people who hate the framework word. I know there's a lot of people who actually hate the word, because I thought that when someone says framework, they're coming to me with kind of like a ready-made structure that this is what you need to do. But I only, like lately, realized that there is a difference. So a difference between a framework and a playbook. So it's like, the thing I don't like is well, okay, i know we both really like American football. So I mean, if there is a playbook game, then that's the playbook game. So I mean you have a move for everything. Like if someone passes this way, you run this way And like so I think data governance can't be that that someone comes to you and says that, oh, i have this model and now you implement this exactly.

Speaker 2 9:22

And when someone says this, you run this way. That's not how you. I mean og nå implementerer du dette, og når noen sier at du ruller på denne måten. Så er det ikke sånn at du jeg mener, selv om du tar en playbook fra en team og du putter det i en annen team, så kommer det ikke til å fungere, for du må byg. Then the playbook is your company's version of implementation of the framework, or how you interpret it, what came out of the framework when you used it. So framework is this kind of scaffolding tool that helps you reveal the best path to success.

Speaker 1 10:01

It's kind of interesting and I read this quite a lot actually, especially on LinkedIn. Someone has posted we've talked to 150 companies and we designed this blueprint to data governance And you click like and you can download it, stuff like that right, but for company number 151 it doesn't work right, because you are trying to build a playbook. It's not a framework. Let me tryøve å tenke på det i en annen perspektiv. Skal du si at en framework har noe i sammenheng med best praktikk? For vi snakker mye om frameworker når vi snakker om metodologi og prosjektmanagerings? agile er en framework. Skal du si det er en forbindelse mellom best praktikk og framework, connection between best practice and framework?

Speaker 2 10:44

It's an interesting question. To me, the framework is a list of questions that you need to answer. I think then, depending on how you answer, then there are best practices. So they are connected, but it's not the same thing to me. I don't know how you see it.

Speaker 1 11:04

I think it's very much connected and you're right, it's not the same thing, because you can, men det er ikke det samme for meg. Jeg vet ikke hvordan du ser det. Jeg tror det er veldig sammenlignet. Du er rett, det er ikke det samme, for du kan bygge din framgang basert på beste praktikk eller du kan bruke, for eksempel, skaldet agile. Det er en beste praktikk, men det er også en framgang som du kan anbefale i ditt organisasjon. When we talk about data governance frameworks, we talk about that very much. How do you get from planning your data governance to adopting? You get that full cycle right. I think that's what makes the big difference between framework and best practice. But then there's another difference And I think this is kind of a nice way ofe fra en hva-det-ikke-nå-perspektiv. En annen forståelse som ofte gjøres og jeg har sett dette mange ganger er at framework og operasjon. Vi har en datagoverningsspramverk og vi har en datamangere-operasjon. Hvordan ser du på forskjellen der? Jeg tror at hvis du anbefaler et, datagoverningsspramverk, i alle fall.

Top-down vs. bottom-up

Speaker 2 12:00

I would kind of think that if you apply a framework, a data governance framework at least the way that I describe it then it will tell you how you should construct your operating model, or it will give you hints to this. I mean, my approach to building data governance in any company is very non-evasive. Building data governance in any company is very non-evasive Not exactly what Mr Steiner's book says, although they are great, but I really love this word non-evasive, because it's like well, like in some sense, with the playbook you try to or with the NFL playbook, you try to identify your company's strengths Who are your best runners, who's your best quarterback And then you play to these strengths and then you kind of build the data governance around existing processes, existing people who are already working with data. So just try to not so that it kind of fits into the culture of the company. Of course, data governance is always a cultural change, but it's like do you want to really, how much change do you want to apply in the beginning, or how much do you expect that you can achieve if you ask everyone to change in the first minute?

Speaker 2 13:19

I think it's kind of like they are again connected, but one leads to another, and then with the operating model, here is where you would apply the best practices. Jeg tror det er noe som de er sammenlignet, men en leder til en annen og med operasjonsmodellet. Her er det der du kan oppføre de beste praktisene, selv om det er en av dine valg.

Speaker 1 13:30

Du har allerede dyvet litt inn i det. Vi prøvde å definere en datagoverning, en framgang fra ulike perspektiver, men du har også tittet på hvordan vi faktisk skaper en datakonsert som er tilgjengelig for vår organisasjon, og kanskje det er noe vi bør diskutere litt før vi går inn på 5L på et høyt nivå. Hva tror du er de største tingene å forstå når du begynner din reise? Jeg tror det første er at vil kompanien virkelig vilje å involvere seg i denne reisen?

Speaker 2 14:04

Fordi det er en reise really want to embark on this journey, because it's a journey, I mean. Often the journey starts with that. There is a kind of revelation from some point that our data is not meeting our needs or we need better, or there's issues quality, I could name like anything and someone would nod and say, yes, we have it. But it's also like it's not enough that you hire a person. You also need to understand that it will require resources and it will require commitment and it will require you to do something. So you can't buy it from outside, but you actually have to. If you want things to be better, you have to do things differently, And that would be kind of like my first thing that do. I think that there is commitment.

Speaker 1 14:54

And that kind of leads us to really the first of your five elements to data governance frame the top-down versus bottom-up. You talked about that commitment. Is that where you should start When, fra topp til nede versus bottom-up? Du snakket om det kommittementet. Er det der du burde begynne?

Speaker 2 15:07

Når en kompani velger å gå inn i data-governance, som jeg sa, det kan være mange ting hvorfor de gjør denne valget. Det kan være en regjering som avviker et veldig høyt nivå som kan bli skjedd av GDPR eller legislasjon eller noen slags kjøp i markedet, at de er som å si at, ok, vi må behandle data som vårt assett. I hvilket fall, hvordan skal jeg si det er kjøp av det fra toppen, fordi det kommer fra to top to down, like the need, and the other direction can be that there are people who are struggling with data and they want to do something better. This can be then that the kind of the need comes from a specific area, for example finance. They usually have a lot of power for well, they have the money, so they got the power.

Speaker 2 16:04

Kommer fra en spesifikk area, for eksempel finans. De har ofte mye kraft for vel, de har penger, så de har kraft. Eller noen andre komplekser eller det er mer nødvendig å forbedre menneskens kraft til å gjøre sin egen rapport. Det kan være noe slik, og dette også hvordan det kommer fra i organisasjonen kan forstå hvordan du vil strukturere ditt datagreff.

Speaker 1 16:35

Ja, dette er et veldig interessant og jeg har alltid følelsen at når det er topp-down og jeg tror du nevnte det, jeg vet ikke om det var på grunn av noe, men når det er topp-down and I think you mentioned it, i don't know if it was on purpose, but when it's top-down it's often compliance-driven That there's a new regulation in the sector. There's kind of an obvious need to be compliant and have a risk-based approach to data handling. It's more natural for them to start from a top-down perspective and say, well, this is what we need to do to be able to have our license to operate. But from the bottom up and this is kind of the interesting one is well, if there's a certain area like finance and control that says, okay, we need to have our data in order, or whatever purpose, that might be compliance again, or whatever purpose it might be compliance again but how does that scale from that business unit to other business units?

Speaker 2 17:32

If I open a little bit about how I think about the layers because we talk about top-down and bottom-up and these terms are actually used again. We could probably spend an hour talking about how we could define these. It would be fun, but let's spare our audience that we can take that offline then spente en uke snakket om hvordan vi kan definere dette. Det ville vært gøy, men la oss spare vår publikum at vi kan ta det off-line. Jeg tenker på når jeg tenker på topp og jeg har snakket om topp-managerings, men jeg tenker faktisk på konseptuelle lager. Så jeg tenker på meninger av ord, av regjering, av definere hvilke typer data vi har. Så du kan gjøre dette gjennom konseptuell modelling. Du kan bruke andre teknikker også, men i og med for meg er topp layer, is the conceptual layer, just very high level. I have here my customer, my transactions, my sales, my locations, my product, my plant. So defining what data we have and then how you want to group this and who owns it and so forth. So just the ownership of the concepts. Then you have like botten. Dette er ditt aktuelle for meg. Som jeg sa, jeg kommer fra tekstsiden, så jeg ser det som aktuelle database-tabler. Så hva du faktisk har tilgjengelig noe, eller rapporter. Det kan også være rapporter eller dataseter tilgjengelige i noen system, så det data som du faktisk har tilgjengelig. Og så har du den midtstilen, som er en slags abstraksjon av det Domain, er en, men jeg forstår også data-produkt, som er i midten mellom det konseptuelle og det fysiske, så den midten. Så hvis vi har noen som kommer fra toppen, the conceptual and the physical layer, so this kind of like middle.

Speaker 2 19:17

So if we have kind of someone approaching top down, then they usually start with that. We want to categorize our data. We want to understand what data each domain uses. We want to assign ownership here. We want to discuss key terms and I actually discussed with one company. it was interesting that they had a data catalog, but for the first I think it was two years they didn't connect it to any system. They didn't bring in any metadata from like system, metadata from like these are our tables. They just described their language and the type of data they had and I was thinking like how did they get the funding for that? but it's so beautiful, that's interesting. I haven't seen that before og den typen av data de hadde.

Speaker 1 19:56

Og jeg tenkte hvordan fikk de fonden for det? Men det er så fint, det er interessant. Jeg har ikke sett det før. Det er fordi av fonden, men en interessant oppgave. Så jeg sier topp og nedgave er fra en konseptuell nivå. Du begynner fra en konseptuell nivå og så drar du deg ned Og i bottom-up be the physical implementation and then try to abstract it from there. I thought it was interesting that you said data products are somewhere in the middle. Would you say logical on logical layer?

Speaker 2 20:23

Yeah, i mean, if you want to think about modeling layers, then logical although that's not again exactly how logical modeling works, but kind of like. I'm thinking this is now a combination of modeling layers and then some conceptual thinking from my side med modellingsverk. Men jeg tenker at dette er nå en kombinasjon av modellingslayer og noe konseptuelt tenkning fra min siden. Men hvis jeg tenker på kjernen av nå opp, ned eller siden eller hva du egentlig vil, hva er den kjede spørsmålet her for denne entiteten? Det er hvilken objekt din datagoverning bør fokusere på.

Speaker 2 20:53

It is that what is kind of the object that your data governance should center around. So like if I, as I said, if you come down, you start doing this conceptualizing and you usually then create the governance around the concept. So the roles are then governing these concepts. If you come bottom up, then you have like tables or datasets is fancier name, but I mean it's in essence, it's a table or a view. So it's a physical thing although we are in the cloud world, but still it's a physical thing to me in the database and you can build the governance around these, like what you actually have and what you offer. And then there is something in between. As said, data product is nowadays the thing that people are leaning towards. You could also do this like a domain is a rather large grouping, but data product is kind of something which is a conceptual thing in the middle, but it connects then the conceptual and the physical, and I mean a lot of data governance is now built around the data product. This is also the approach that we're doing at Relaks.

Speaker 1 22:03

There's one question that is a bit open and this goes both ways for the top-down and the bottom-up and that's about the who. who is responsible? So who does actually do the work?

Speaker 2 22:15

You mean the actual work other than PowerPoints and meetings? Yeah, that then depends on how you define your data governance model. I mean, how I think about it is that I see the data governance roles as representing work that needs to be done, and the person who holds the role is responsible for making sure that the work is done, som representerer arbeid som må gjøres, Og personen som holder rollen er ansvarlig for å sikre at arbeidet er gjort, men de er ikke nødvendigvis personen som gjør arbeidet, så bare at de må sikre at noen gjør det. Jeg vet at det er ok. Igjen, jeg håper du kan snakke mye om data steward. Hva er det? Er det nødvendig? Men jeg tror at dette er en kompani-level-diskusjon. Vil du virkelig bestemme deg hele tida til arbeidet eller vil du bare bestemvarlig for å sikre at arbeidet er gjort? Jeg prefererer å stå på ansvarsnivået og ikke gå så langt, men ja, du kan også gå videre ned i denne hulken.

Speaker 1 23:31

Den grunnen jeg spørte om er at vi diskuterte forskjellenom framverk og operasjonsmodell. Vår spørsmål er om du velger en opp og ned og understøtte oppgave, vil det ha en direkte effekt eller behov på hvilken operasjon du kan velge? Hvis du velger en opp og nedstøtte oppgave, er det ikke konseptuelt at du må ha en hå holistisk forståelse av organisasjonen? Hvordan skaper jeg det, Hvordan operasjonaliserer jeg det fra der Og så begynner du fra en sentral hold.

Speaker 2 24:02

Ja, det er en annen veldig god spørsmål. I alle fall, jeg tenker, som jeg snakker, så bør du være med meg litt Når du gjør din datagjørelseplan, hvordan du vil fortsette med din datagjørelse, det vil definitivt se annerledes ut uten at du begynner med at ditt sentrale ting er et dataset eller at det er et produkt som også kan være representert av et dataset, men jeg sier fortsatt at det er konseptuelt større eller om det er et virkelig konseptuelt asett som du regerer. For meg må data regeringen i slutten være i stedet til å holde alle disse entitene. Så det må være konseptuelle modeller, det må være noen slags midterr og så må det være link til de virkelige tingene som er på offer.

Speaker 2 25:12

And you want to construct that with your governance work, but you have to start from some point and then think how you're going to do it Now. Does this dictate operating model? It has to affect it. Like if I'm thinking logically, there has to be differences, but I can't thinkikke regler for dette. Men om jeg bare logisk tenker at det må være en forskjell mellom operasjonsmodellen, hvordan du begynner å strukturere det og hva er din kjørestruktur.

Defensive vs. offensive Data Strategy

Speaker 1 25:48

Det vil sikkert være interessant å utforske. Jeg tror kanskje det er litt vanskelig å si at det krever en certain operasjonsmodell, men du har rett, jeg tror det er en korrelasjon. La oss snakke om den andre elementen litt, og jeg tror dette er en veldig interessant fordi jeg har snakket om det med fotballer kjøring av defensiv eller offensiv. Jeg har lurt en artikkel fra Thomas Davenport i 2017, jeg tror det came out where you talked about what's your data strategy? right, choose a defensive strategy, offensive strategy. Align that with your organization. So if you work in a highly regulated industry, it might be smart to choose a bit of a defensive strategy. If you work in an industry like retail, for example, offensive strategy can help you advance very much. But is that kind of the direction you are going with the element of defense versus offense in your data governance framework?

Speaker 2 26:44

I know that this is a topic that you have spoken and written about, so I mean, i agree with you fully that in some sense, it's case by case which strategy you take, so it's not like one size fits all. again, i think what I'm more like leaning towards this question. as I said, with the top-down, bottom-up, the key question to answer is what is the entity that you want to build your data governance around? so what is the governing entity? I think, for this defense or offense, my kind of key question, or what you want to answer is why are you like? what is your motivation as a company for starting to build the data governance?

Speaker 2 27:28

Is it like, as we said, if it comes from the management, it is often some kind of fear or risk mitigation type of motivation, in which Det er ofte en slags fyr, eller risikomitigasjonsmotivasjon. i hvilket fall, vil du begynne med ting hvor du kan vise at vi nå er, på en måte, med å mitigere risikoen? Og hvis det kommer fra en slags sted for å åpne, gjøre tilgjengelig, virkelig vilje til get more use out of data, wanting to be more data-driven, then you would start with use cases that are more in line with this, where you can show that okay, now here we are, achieving this, so it's more like setting a target more than choosing like exactly how should one put it? exactly like this is how we will always be. There are, of course, sector like sector driven differences, that the banking and other sectors that are highly regulated tend to lean more in this defense, which is natural.

Speaker 2 28:35

If you don't get things right, you go to prison. So there's your motivation. But yeah, that's what I'm looking for with this question. is that what is the motivation, like the highest motivation on the company level, to go for this?

Speaker 1 28:49

And I think that has a direct effect on what you are prioritizing in your activity. I think, if you were defensive, i'd rather look into how do I standardize my data, how do I ensure standardized access to data storage right? And if you look at offensive side, it's more about how to optimize for value creation through data, how to optimize for analytics, how to optimize for visualization for and maybe I'm mistaken here. So I'm interested to see perspective on that. But I don't think it's black and white. I think there's a spectrum from defensive to offensive. To find your place on that spectrum as an organization. Yeah, definitely It's a spectrum, and it's also fra defensiv til offensiv for å finne ditt sted på det spektrumet som organisasjon.

Speaker 2 29:29

Ja, det er definitivt et spektrum og det er ikke at du bare gjør en ting. Du har brukerheter hvor du vil gjøre data mer tilgjengelig og du har brukerheter hvor du virkelig vil gjøre sikkerhet at ingen bruker dette dataet eller noe uten min formissing og glede. No one uses this data or anything without my permission and blessing. And this high stack of papers signed sickle papers. So, yeah, it's a realm, but kind of like, if we're thinking about, like I used, that this is a framework that it used to start building a data governance, then it gives you the starting point. So, again, this will change en datagoverning, så gir det deg begynnelsen. Så, igjen, dette vil forandre.

Speaker 2 30:10

Men det er hvor man begynner å vise verdien, for det er superkritisk at hvis du får kjøpt kjøp, du får budgetet og du får tilgjengeligheten til å gå foran med dette, så må du være able til å vise verdien. Jeg har funnet dette til å være utfordrende fra tid til tid, for det kan være vanskelig å vise, som det ofte snakkes om tidssavninger eller å kunne gjøre noe mer, men disse er veldig vanskelig å monetisere. I en kompani-setting ser du ofte på eurosigner eller dollarsignere eller, for oss, hero signs that what you have achieved with this. So I think that if there is this kind of commonly agreed that these are the issues that we want to mitigate, or these are the things that we want to achieve, then it's more easy for you, as the person heading this initiative, to show that, okay, i'm achieving my goal, so we can move forward.

Speaker 1 31:12

What I think is really interesting on that defensive offensive, but also maybe on the next element, which is the scope of your data governance work, is the influence of AI and automation. We've seen, and I think each and every company I've talked to, av AI og automasjon. Vi ser nå, og jeg tror hver og hver sammenheng har snakket med podcaster eller andre som har snakket om hvordan vi kan bruke automasjon til å forbedre effektiviteten i vårt datamanagement datagoverningssverk. Det er en side, men den andre siden er også hvordan vi må bruke datamanagement-pratiser og få datagover, use data management practices and get data governance ready to become a motor for automating our business right To ensure that we have the good foundation. Do you see that impacting the defensive-offensive element?

Speaker 2 31:58

Yeah, i mean, i actually just started thinking that if the target would be that the company wants to be more ready for AI or be more or have, i guess, have more, more ready for AI, or be more or have, i guess, have more data ready for AI because that's the core thing when it comes to AI, that you have the suitable data and you have enough of it and it has the metadata around it I was actually thinking that is that. Would I categorize that as offensive or defensive? I'm not actually sure. It's a bit both, because in a sense, if you consider that this is a capability, that will be a competitive advantage and actually later, if you don't have it, it could be a disadvantage competitively. So you could say That it's a defensive strategy, that you defend your, you know your market share Or whatever. Or you could consider that it's This Attack play that we are now. We want to be better than others. So I would say It comes down. It comes actually Down all the way to your strategy That.

Speaker 2 33:09

How do you see your company's strategy That? are you the forerunner? så jeg vil si at det kommer faktisk ned til strategien din. Hvordan ser du kompaniens strategi? er du forståelsesrunner? forstår du deg som å skrive nye ting og kompaniens å gjøre nye produkter, nye forståelser, eller er kompaniens strategi å produsere stabile byggene bedre og mer effektivt enn dine foreninger? Så du kan naturligvis bruke AI i å oppnå hver eneste av disse, men jeg tror jeg at jeg ikke vil plassere AI i hver eneste av disse sektorerne direkte. Det beror på hva du vil gjøre med det.

Define the right scope & stakeholders

Speaker 1 33:46

Veldig god ansvar. Jeg tror vi kan ta litt av det til neste element også, når vi snakker om skåp, hvordan reddere du rett skåp for ditt datagrepsverk?

Speaker 2 33:59

Skåp er som om det er vanlig. Hvis du sier de to første kjene, så vil disse faktisk guide både skåpet og dine k the two first keys then this will actually guide both your scope and your key stakeholders. So this is kind of like they are feeding. Especially this defense offense, or the motivation will feed into this. Then you have to see, like, if you think about scope and stakeholders, they are naturally tied together to some extent, depending on what you choose. Of course, with the stakeholders, there's always the IT business question which rises. But if we think about the scope, then your choices in the beginning will probably identify which area it is that you want to continue with. I'll go into the stakeholder bit because I already touched it. So then, if you know which is the kind of the area in the business that you want to work with, then this is where you should find your stakeholders, and I would vote for business stakeholders if I had to choose.

Speaker 2 35:05

This was also like a question posed to someone in the I don't remember who it was, but it was in the CD or IQ in Helsinki that if they had to choose, would they have like business or IT, and they said that if they really had to choose, they would go for business Again. You can have different and it depends also on the person, but I would say that business needs to drive det. Det er den veien. Hvis du vil ha business-valg, så må du ha business-initiativ. Hvis du vil ha kostsavings i noen sektorer, i IT, så IT. Men vanligvis er regjering ikke en kostsavingsaktivitet som en sånn, not a cost-saving activity as such. It can have those effects, yes, agreed, but I don't consider it a cost-saving as such.

Speaker 1 35:58

But I really like that you not only connected scope to the question of defensive versus offensive and top-down versus bottom-up, but also key stakeholders, because this is very much connected right om defensiv versus offensiv og topp-down versus bottom-up, men også om kjøp-stakeholdere, for dette er veldig sammenlignet, og jeg gjorde en utdannelse i årsbøk om hva som er det vi beskrever som kjøp og business-nødene fra kostsavning til tilkomplejens, til reputasjon, til forholdsutvikling og to reputation to revenue growth and then understand how can we connect that to our data governance work. And you're right, you have to find out what fits with your organization And if your organization is very much cost-driven, then that's something you have to connect. Also on the stakeholder side, let's talk about the final element, which I think is kind of interesting, because the entire episodeordet med Juha var om datakonsultanser og hvilken rolle konsultanter og eksternalere kan ta i data og organisasjoner. Hvordan ser du på rollen av eksternalere fra en regjeringsperspektiv?

Speaker 2 37:04

Jeg putter denne kjøret her fordi av mine personlige opplevelser. Dette er ikke til å si at jeg har dårlige opplevelser med konsulter Mange av mine beste vennler er konsulter for realitet. Men det er mer sånn at det er vanskelig å finne konsulter som du kan taage i businessen, som kan selge businessen direkte. Data governance er også en selskap Jeg bruker det dårlige ordet som ingen bruker i IT som er å selge. Det er det det er. Selv om man har mandatet, må man fortsatt bevinne partiene at de utf, de gjør forhold, de gjør forandringer og de får tilgjengelighet. Og denne selskapen, føler jeg, du kan ikke utsource det. Det er annerledes når noen som er i samfunnet, din samarbeider, kommer og forteller deg noe. versus en konsultant. Det burde ikke være, men dette er bare sånn det er. versus a consultant. It shouldn't be necessarily, but this is just how it is And I've seen this in some projects that they want to hire a consultant to drive the data governance project and I don't think that's a good idea.

Speaker 2 38:20

There has to be a face inside to you know the face of this project. Yes, an actual face, not a team, but a face. The other thing is that it can be difficult to also i denne prosjekten, ja, i en faktisk fase, ikke en team, men en fase. Det andre er at det kan være vanskelig å finne, for eksempel, en datakatalogtul utenfor denne toolen. de større i dag har en konsultansbasis som de stiger. Men også hvis du er i et små nordisk land så kan det ikke være så bredt det du har tilgjengelig. they are growing, but also if you're in a small Nordic country it might not be that wide what you have available. so kind of like if you're looking at tooling again in this consultant kind of framework, think like that that what also consider when you're buying the tool, that do a little bit of outsourcing, that can you actually find help.

Speaker 2 39:06

my own experience was that if you do find consultants because they come and go anyway that's their nature outsourcing at du faktisk kan finne hjelp. Min egen opplevelse var at hvis du finner konsulter fordi de kommer og går det er deres natur de forandrer projekter. så dette fortsettelse må du sikre at det er der så du ikke får inn problem med din teknikk, fordi du vanligvis har en slags hjelp som du også vil bruke som din datagjørelsehjelp, because usually you do have some kind of tool that you want to use also as your data governance tool and to hold all of these roles, plus then all of the tech stuff and the fancy metadata, but also like so, this, like that how do you want to use the consultants? make sure that there is that kind of people available, there is continuity. Is that kind of people available? There is continuity. And also, like when you are discussing with potential, interviewing potential people, then try to see if they think like you, Because some, if they have very like people have different experiences and naturally you always bring your experiences to the table. I mean, that's what you do as a consultant. You have experiences, you have skills.

Speaker 2 40:08

Selvfølgelig bringer du alltid dine opplevelser til bordet. Det er det du gjør som konsultant. Du har opplevelser, du har kunnskaper og så bringer du disse til klienten. Men hvis du tenker metodologisk veldig annerledes, så kan dette også være et problem. Så jeg tenker på dette holistisk. Så igjen, la oss gå med playbook-sens. Så kjøp vyselig dine partnerer og tenk på dette også strategisk. Tenk ikke at jeg bare kan høre noen. Jeg tror det er det som er det største messet her.

Speaker 1 40:42

Det er et veldig godt messet og jeg tror det er veldig mye en forskjell av å få good message. I think there is very much a difference of getting externals into your organization the way you described it and just trying to fill some kind of resource gap or manning up on project, because that's also an approach that some organizations choose. We have to have a project for a certain period of time. Let's man up and find more hats on their project. But I think you're right. I think you talked about new perspectives that come in through consulting. You talked about that specialty knowledge that you can get. There's also something about pattern recognition. They can see certain key learnings they had from their experience from other organizations and bring with them. De kan se nødvendige læringer de har fra deres opplevelse fra andre organisasjoner og bringe det til organisasjoner. Og noe som er en veldig viktig rolle av en konserter er å være en skapkøt for alt som går feil. Man kan også alltid skade det på konserter.

Speaker 2 41:39

Jeg håper ikke Min egen oppgave til dette har alltid vært at. Hvis jeg er data manager, in any case, i make the final decisions, so I will follow with my decisions. So I mean that's how I think about it. I will take advice and sometimes the advice will lead to good results and sometimes it will not, but I mean, the final decision is always mine, so I protect my own. But that was now a personal note. But even if I spoke about being strategic with this, then I think it's very good that you emphasize this, that there is value there, there is learnings there.

Speaker 2 42:23

As said, there is very often people who are the first of their kind I'm actually the first data governance manager Alexa's ever had er det ofte folk som er første av deres? Jeg er faktisk den første regjeringsmanageren Alexa har hatt. Så la oss se om jeg er siste av deres. Men i en måte kan det være litt løgnlig Og i dette igjen vil jeg komme tilbake til det vi snakket om i begynnelsen at kommuner som DEMA kan være av bruk her. Du kan finne der dine terapie-menn som du kan dele dine behov med, og det er også viktig at du kan diskutere om dette.

Speaker 2 42:59

Det gode med datagoverning er at det ikke er kjøret av en kompaniens business, så du kan begynne å forhåpentlig gjenstille informasjon. It's not the core of a company's business, so you can usually quite freely exchange information about how you are building your data governance. You can't talk about the specifics of what data you are maybe governing, but you can talk about how you're structuring the models. Usually, there is not a problem sharing this information from a proprietary perspective. So that is the good thing about this area that you can be quite freely network and have this kind of exchange and support from the community.

Speaker 1 43:37

Fantastic, and you framed the entire organization perfectly, so I almost don't have to ask about key takeaways and call to action. but if there's one thing you would say, this is what I want you to do now, after you dive into data governance framework as five elements, what would be the one thing you would recommend?

Speaker 2 44:00

I would maybe start with that. I would hope that if you look at the five elements and now listen to us discussing this topic, i hope you would not be so afraid at hvis du ser på de fem elementene og nå lurer du på oss om det er det vi snakker om, så håper jeg at du ikke blir så bekymret, for jeg føler at datagoverning er noe slik en skjønnelig topp. Det er forstått som å være veldig vanskelig, ikke at det er lett, men også at du ikke bør bekymre deg. Det er bedrifter der, så du bør virke fear it. There are benefits there. So you should really like if you have problems, then try to solve them.

Takeaways and Call to action

Speaker 2 44:32

Sometimes the cure is worse than the disease, but I mean, i do consider this data governance a cure. This is why I've ended up in the area. So I think that that would be kind of my thing, that if also, if you have data governance already existing in your company but you're not getting the benefits out of it, then don't be afraid to say it and try to fix it, Because at least that's how I see my role, that I want to make things better, and data governance should make things better, except in the regulatory context, where it should make everything difficult and horrible. Okay, okay, sorry, but yeah, i would want to bring you hope that things can be better with your data.

Speaker 1 45:21

Fantastic to end on To hope. Thank you, Sadi.