Malt’s data ROI framework to prove your data team’s worth in big, bold numbers

Tarush Aggarwal (00:00)
Welcome to another episode of People of Data, where we get to highlight the wonderful people of data. These are our data leaders from various industries, geographies, and disciplines. Discover how companies are using data, their achievements, and the challenges they face.

Today on the show we have Anais Ghelfi who is the head of data platform at Malt. Malt is, as we all know, a leading freelance marketplace in Europe and the Middle East. Anais oversees data engineering, analytics, ML ops. In prior experiences, she spent five years at...

Pernod Ricard where she built the company's data platform and led the implementation of the data transformation initiatives. She's also a strong advocate for women in tech and diversity in the workplace. Welcome to the show Anais. Thank you for having me. Of course, thank you for being here. You know, very excited about diving in. So are you ready? Yeah, let's do it. Great, let's do it. So, you know.

‍

Anaïs Ghelfi (00:45)
Thank you for having me.

Yeah, let's do it.

‍

Tarush Aggarwal (00:59)
What does Malt do? How does Malt make money? How does data play a role in helping Malt do what it wants to do?

Yeah, that's big question. So first of all, MALT is the leading freelance marketplace in Europe. So on MALT we have over 70,000 of companies at all sites, so from small to enterprise businesses that find the talent they need in MALT's community of more than 700,000 freelancers today. The core of our marketplace is really our matching algorithm.

‍

Anaïs Ghelfi (01:07)
Yeah, that's a big question. So first of all, MALT is the leading freelance marketplace in Europe. So on MALT, we have over 70,000 of companies at all sizes, so from small to enterprise businesses that find the external talents they need in MALT's community of more than 700,000 freelancers today. The core of our marketplace is really our matching algorithm. Our end goal is to find the perfect match between a client's project

‍

Tarush Aggarwal (01:33)
Our end goal is to find a perfect match between a client project and a freelancer.

‍

Anaïs Ghelfi (01:37)
and a freelancer. And as you can imagine, in a company like that, which is a pure tech company, a data team has a really central role in the company. So first of all, we directly contribute to the product itself because we develop and evolve everything related to matching algorithm, which is the heart of the product, and also developing features that are using AI. So there is really a first contribution on the product itself. Second part, as a data team, they are

‍

Tarush Aggarwal (01:38)
And as you can imagine, in a company like that, which is a pure tech company, a data team has a really central role in the company. So first of all, we directly contribute to the product itself because we develop and evolve everything related to matching algorithm, which is the hub of the product and also developing features that are using AI. So there is really a first contribution on the product itself. Second part, as a data team, they are...

‍

Anaïs Ghelfi (02:05)
Analyst team, you know, that are really translating company strategy into measurable KPI. So then you have everything related to analytics, reporting data, and so on. And the last part I'll say is we do build a data platform to enable all the data projects. And we built on top of that what I'm calling data products. So it's really like specific products to answer a specific business need. So if I give you one example, or maybe two examples.

‍

Tarush Aggarwal (02:05)
you know that are really translating company strategy into measurable KPI. So then you have everything related to analytics, reporting data warehouse and so on. And the last part I say is we do build a data platform to enable all the data projects.

And we built on top of that what I'm calling data products. So it's really like specific products to answer a specific business need. So if I give you one example, or maybe two examples, we built an internal application that enables anyone in the company to use Gen .AI. So you ask a question about the product, you have to answer. There are many, many other use cases. So that's one example. Another one would be that we are building a custom data products to...

‍

Anaïs Ghelfi (02:31)
We built an internal application that enables anyone in the company to use Gen .AI. So you ask a question about the product, you have the answer. There are many, many other use cases. So that's one example. Another one would be that we are building a custom data products to help the sales team generate leads. So of course, that requires enrichment of data, collection of data and so on. So I say really that this data team plays different role in the company, but we do support a lot of part of the business as of today.

‍

Tarush Aggarwal (02:47)
help the 16 generic leads, so of course that requires enrichment of data, collection of data and so on. So I say really that this data team plays different role in the company, but we do support a lot of part of the business as of today. So exciting, I love that.

One of the applications of what you are doing is fundamentally contributing back to what's product because on the matching and algorithm side, so we would love to sort of, I would love to dive into that deeper as we get later on in the show. What I would love to know is, you know, how does, what's the culture like at Malt? And you know, what's the overall macro culture you see in the company and how does this really translate to the culture which you're building inside the data and analytics group?

‍

Anaïs Ghelfi (03:33)
Yeah, I think we really have a culture. So we are a tech company. So Malt has been built 10 years ago by two person, one guy who was a CTO. So he was really building the platform and he was freelancer itself. And then another guy that was, you know, really leading the vision and the product. So having said that, the culture is really built on top tech and product with a culture that really is focusing on craft, innovation.

‍

Tarush Aggarwal (03:33)
Yeah, I think we really have a culture. So we are a tech company. So Maltz has been built 10 years ago by two person, one guy who was a CTO. So he was really building the platform and he was freelancer itself. And then another guy that was, you know, really leading the vision and the product. So having said that, the culture is really built on top of tech and product with a culture that really is focusing on craft, innovation,

‍

Anaïs Ghelfi (04:03)
empowering team and so on. And I think as of today, we really still have this culture where we are a tech company, we can innovate, engineer and products. It is at the heart of our business. So we are really building on top of that. And then when it comes to a data team, with of course, you can imagine when you build a company, you don't have a data team 10 years ago. And this is something that is coming after you have a few product analysts that is helping, you know, helping product manager to find insight and then

‍

Tarush Aggarwal (04:03)
and empowering team and phone. And I think as of today, we really still have this culture where we are a tech company, can innovate, engineer and products. It is at the heart of our business. So we are really building on top of that. And then when it comes to a data team, which of course you can imagine when you build a company, you don't have a data team 10 years ago. And this is something that is coming after you have a few product analysis that is helping, you know, helping product manager to find insight. And then you end up having a bigger data team that is

‍

Anaïs Ghelfi (04:31)
you end up having a bigger data team that is building a platform, that is building data warehouse, product matching algorithm and so on. And I'd say that in the data team, the culture is very much similar to the culture of the founders, because we are really focusing on craft innovation, so it's something really important for us. And also making sure we can take risk and fail fast. So this is also something we really have in the team. So I'll say that the data team culture is basically like,

‍

Tarush Aggarwal (04:33)
building platform, that is building data warehouse, product matching algorithm and so on. And I say that in the data team, the culture is very much similar to the culture of the founders, because we are really focusing on craft innovation, so it's something really important for us. And also making sure we can take risk and fail fast.

So this is also something we really have in the team. So I say that the data team culture is basically like an extent of the culture of the company, which is mainly focusing on tech and product. What's one aspect of the data team culture which is different, you know, like, you know, we see a lot of companies where the culture of the founders are top down is really what sort of permeates. But at the same time, teams have their own individuality and their own quirks in the way of doing things.

‍

Anaïs Ghelfi (05:00)
an extent of the culture of the company, which is mainly focusing on tech and product.

‍

Tarush Aggarwal (05:22)
I would love to hear what's one of those quirks which the data team at Malt has. Does that make sense? Yeah, yeah, it makes sense.

‍

Anaïs Ghelfi (05:32)
Yeah, it makes sense. So as a data team, and as I explained a bit earlier, we are really working with a lot of things, a lot of teams, so supporting a different part of the business. And I think this is also the strength of our team is to be as traversal, meaning you work with different people and different businesses in the organization. So you have a transversal view that allow you, if I come back to the culture, talking about innovation, for instance, allow you to see, okay, here we can improve stuff.

‍

Tarush Aggarwal (05:34)
So as a data team, and as I explained a bit earlier, we are really working with a lot of things, lot of things, so supporting a different part of the business. And I think this is also the strength of our team is to be as transversal, meaning you work with different people and different businesses in the organization. So you have a transversal view that allows you, if I come back to the culture, talking about innovation, for instance, allow you to see, okay, here we can improve stuff. And I can give you an example later if you are interested.

‍

Anaïs Ghelfi (06:01)
And I can give you an example later if you are interested in. Coming back to the product, we see that we'd like to improve our internal efficiency because we see there are a lot of manual process, know, the business is scaling, we still have manual processes, we are not efficient, we need a lot of people to do stuff. And when you look at that as a data team, because we are engineers, we are like, okay, we can do something about that. So if I take the example of internal efficiency,

‍

Tarush Aggarwal (06:05)
Coming back to the product, see that we'd like to improve our internal efficiency because we see there are a lot of manual process, the business is scaling, we still have manual processes, are not efficient, we lot of people to do stuff. And when you look at that as a data field, because we are engineers, we are like, okay, we can do something about that. So if I take the example of internal efficiency, something that we can do is like, for instance, earlier this

‍

Anaïs Ghelfi (06:30)
Something that we can do is like, for instance, earlier this year, we came up with a project saying, we do believe that Gen .AI could help all our teams being more efficient in a company. We pitched that, people say, okay, just try, let's see if it works. So we tried a POC earlier this year, and now we have a product that is fully developed, available to all the companies and used by all the employees internally. And I think it's a good example. And then we can come back to that.

‍

Tarush Aggarwal (06:33)
Here we come up with a project saying we do believe that the NIA could help.

all our teams being more efficient in a company. We picked that, people they say, okay, just try, let's see if it works. So we tried the POC on this year, and now we have a product that is fully developed, available to all the companies and used by all the employees internally. And I think it's a good example, and then we can come back to that later. I think it's a great example of the product we deliver in the data team. Looking at that, it really translates the culture is like, how?

‍

Anaïs Ghelfi (06:59)
later, I think it's a great example of the project we deliver in the data team. Looking at that, it really translates the culture is like how to be innovative. It's not a top-down approach. It's something really bottom-up that you identify, and then the culture lets you do that. So I think that's how I will explain how the values can be translated in the day-to-day.

‍

Tarush Aggarwal (07:07)
to be innovative. It's not a stop down approach. It's something really bottom up that you don't see side. And then the culture lets you do that. So I think that's yeah, how I will explain, know, how the values can be translated in the day to day. I'm super interested in the Gen .AI use case which you have. What's an example which somebody would go ask, you know, off this product on a day to day basis?

‍

Anaïs Ghelfi (07:31)
Yeah, so this project started with a first pain point, which is when you work at Malt, you know the product is expanding very fast. We have a lot of new features and all the team cannot keep up with that because the content is spread on Google Drive, Notion, Slack, and so on. And we have a culture where you just ask your colleague in Slack and everybody's losing time. So we started from them and say, okay, we have a product knowledge base that is kind of well structured.

‍

Tarush Aggarwal (07:31)
Yeah, so this project started with a first thing point, which is when you work at Mald, you know the product is expanding very fast. We have a lot of new features and all the team can keep up with that because the content is spread on Google Drive, Notion, Slack and so on. And we have a culture where you just ask your colleague in Slack and everybody's losing time. So we started from them and say, OK, we have a product knowledge base that is kind of well structured.

‍

Anaïs Ghelfi (08:01)
on different people on the product side, on the customer care side. And we built a and a bot that we put in Slack where anyone in the company can come and ask a question. So I'll give you an example. You know, at Malt, we are working with different countries. So you can be like, okay, if I have a freelance service in Belgium, can you work for a French company? And this type of things like that. And basically, this project is really building the technical stack.

‍

Tarush Aggarwal (08:00)
and back to different people on the product side, on the customer care side. And we built a rag and a boat that we put in Slack where anyone in the company can come and ask a question. So I'll give you an example. You know, at Malt, we are working with different countries. So you can be like, okay, if I have a freelancer based in Belgium, can he work for a French company? And this type of things like that. And basically this project is really building the technical stack.

‍

Anaïs Ghelfi (08:30)
embarking the right people because they know the knowledge much better than we do. And then we build that. We open the tool to 15 people and now we have 400 people using it as of today. So here is really building the capabilities which is using Gen .AI and leading a project that is bringing value. And when you have one use case that is working, you can just expand to many, many use cases. And as of today for this project, we have more than 10 use

‍

Tarush Aggarwal (08:30)
embarking the right people because they know the knowledge much better than we do. And then we built that. We opened the tool to 15 people and now we have 400 people using it.

as of today. So here is really building the capabilities which is using Gen .AI and leading a project that is bringing value. And when you have one use case that is working, you can just expand to many, many use cases. And as of today for this project, we have more than 10 use cases available in the company. That's awesome. Talking about...

‍

Anaïs Ghelfi (09:00)
10 use cases, sorry, available in the company.

‍

Tarush Aggarwal (09:06)
data and your data team, what does the data stack look like? What does the data team look like? How big is the company? How big is the data team? What's the breakup? Yes, so as a baseline at Malt, we are between 500 and 600 people, like employee working mostly in Europe. Looking at the data team, it's between 25 and 30 people.

‍

Anaïs Ghelfi (09:15)
so as a baseline at Malt are between 500 and 600 people, like employee working mostly in Europe. Looking at the data team, it's between 25 and 30 people. In the data team, have three, I'll say three big teams. The first one is the data analyst team. So like, why you have all the data analysts working with the business and the product. The second team is the data science team.

‍

Tarush Aggarwal (09:30)
In the data team, you have three big teams. The first one is the data analyst team, so like where you have all the data analysts working with the business and the product. The second team is the data science team, where you have the machine learning engineers and the research engineers. And the last one is the data platform where you can find the data engineers, analytics engineers, and ML ops. yeah.

‍

Anaïs Ghelfi (09:43)
where you have the machine learning engineers and the research engineers. And the last one is the data platform team, where you can find the data engineers, analytics engineers, and MLOps.

‍

Tarush Aggarwal (09:56)
And in the 25 person data team, you said how big would your data platform team be versus the AOML team versus the analytics team?

‍

Anaïs Ghelfi (10:08)
Yeah, if I think of the number of people, so the data platform team, it's four data engineers, three analytics engineers, and two MLOps. So it's really all these people that are building the foundations to make everything available for all the other team. In the data science team, you have five person and data analysis between five and seven. So the ratio is quite spread across the different teams.

‍

Tarush Aggarwal (10:08)
Yeah, if I think of the number of people, so the data platform team, it's four data engineers, three analytics engineers, and two ML ops. So it's really all these people that are building the foundations to make everything available for all the other teams. In the data science team, you have five persons and data analysis between five and seven. So the ratio is quite spread across the different teams. Yeah. And what does the data platform look like today? What are some of the tools?

Yeah, so the technical stack basically is on top of Google Cloud Services, so Google Cloud Platform. I think it's a mix of Google Cloud Services. And if I give you a few examples, the backbone of the data warehouse is on BigQuery. Orchestration, you have Composer, which is the airflow managed by GCP.

‍

Anaïs Ghelfi (10:39)
Yeah, so the technical stack basically is on top of Google Cloud Services, so Google Cloud Platform. I'll say it's a mix of Google Cloud Services. And if I give you a few examples, the backbone, the data warehouse is on BigQuery. Orchestration, you have Composer, which is the airflow managed by GCP. And then when it comes to reporting, we have Looker. So here are the three main components.

‍

Tarush Aggarwal (11:03)
And then when it comes to reporting, have Luca. So here you know the three main components. But we do rely and deploy a lot of open source tools. To give you a few examples, we are using MLflow for the models. We are using Qdrant as a vector database. We are using Lengfuse as the LLM monitoring tool. And then today, when it comes to also data stack and transformation, we have an in-house tool that we developed.

‍

Anaïs Ghelfi (11:08)
But we do rely and deploy a lot of open source tools. To give you a few examples, we are using MLflow for the models. We are using Qdrant as a vector database. We are using Langfuse as an LLM monitoring tool. And then today, when it comes to also the data stack and transformation, we have an in-house tool that we developed and we are looking at integrating dbt or SQL mesh, for instance.

‍

Tarush Aggarwal (11:33)
and we are looking at integrating Dbc or SQL mesh principles. But it's basically Google Cloud plus open source services with deployed Yeah. How about data ingestion? How are you the data into BigQuery, especially for the things outside of GCP?

‍

Anaïs Ghelfi (11:37)
But it's basically Google Cloud plus open source services we deployed

Yeah, so we are using two tools today. So we have Airbytes, which is an open source tool that manages most of our data pipeline. And for some other pipelines, we have Fivetran. So it's Fivetran for some pipelines and Airbytes for, I'll say, 90% of the others.

‍

Tarush Aggarwal (11:53)
Two tools today. So we have AirBytes, which is an open source tool that manages most of our data pipeline. And for some other pipelines, have 5Tran. it's amazing.

5 trans for some part-time and 4, I'll say 90% of the others. Awesome, very exciting. What is, you know, and I think you already kind of tackled on this before in some ways with the LLM project, I'm sorry with the Gen AI project, but what is one achievement which, you know, as a team you're very proud of? Yeah, so I have to...

‍

Anaïs Ghelfi (12:26)
Yeah, so I have two. Sorry, I'm cheating a bit. I say one of the biggest achievements is really building a data product. It's not only like doing a project and then leaving it. are really thinking building it as a product, who are the person as we have internal communication. So we did a logo with the marketing.

‍

Tarush Aggarwal (12:29)
Sorry, I'm I say one of the biggest achievements is really building a data product.

It's only like doing a project and then leaving this. are really thinking building it as a product. Who are the best on us? We have internal communication. So we did a logo with the marketing song. I mean, we really like package everything into a product with a clear purpose. So everybody in the company knows what's the name, what it does. I think that's one of the big achievements. I think it's really important, especially when you work in the data platform team, to have your well understood by the company.

‍

Anaïs Ghelfi (12:48)
song. I mean, we really like package everything into a product with a clear purpose. So everybody in the company knows what's the name, what it does. I think that's one of the big achievements. I think it's really important, especially when you work in the data platform team, to have your work well understood by the company. So I'll say that's one first achievement. And the second part is the technical achievement because to build that, technically speaking, it's a you need MLOps, you need data engineering.

‍

Tarush Aggarwal (13:06)
So I'll say that's one first achievement. And the second part is the technical achievement because to that, technically speaking, you need MNLabs, need data engineering. So I'd say it's a mix of different skills. And also when you're using LLM, everything is new, so we tested new tools. And I think what we built is really something that is solid, robust, and can scale to...

‍

Anaïs Ghelfi (13:18)
So I'd say it's a mix of different skills and also when you're using LLM, everything is new, so we tested new tools. And I think what we built is really something that is solid, robust and can scale to almost unlimited use cases. And technically speaking, when you build something like that that can scale, it's also something like the team can be proud of and I'm particularly proud of.

‍

Tarush Aggarwal (13:32)
and almost unlimited use cases. And technically speaking, when you build something like that at scan scale, it's also something like the team can be proud of and I'm particularly proud of.

And if you had to go quantify the ROI of these projects, how would you measure that today? Because I know that's another area which you're super proud of.

‍

Anaïs Ghelfi (13:52)
Yeah, that's a super interesting question because everybody is asking you, okay, how much should we invest? Like what's the ROI? And it's really, really complex to calculate the ROI when you use Gen .ai for knowledge. Because how do you measure that someone is not asking a question directly on stack to a colleague and is using your tool instead? It's really, really hard to measure. So...

‍

Tarush Aggarwal (13:52)
Yeah, that's a super interesting question because everybody is asking you, okay, how much should we invest? Like what's the ROI? And it's really, really complex to calculate the ROI when you use GNI for knowledge. Because how do you measure that someone is not...

asking your questions directly on stack to colleague and he's using your tool in stack. It's really really hard to measure. So what we did in that case, we focused on another use case where we could show the ROI. I'll give you an example. we work with the same products with the customer care team to help them draft the answer of the tickets we are receiving on the support. So this tool is also drafting answer for our clients and freelancer. And here the ROI is much

‍

Anaïs Ghelfi (14:17)
What we did in that case, we focused on another use case where we could show the ROI. I'll give you an example. we work with the same products, with the customer care team to help them draft the answer of the tickets we are receiving on the support. So this tool is also drafting answer for our clients and freelancer. And here the ROI is much easier to calculate because you have the average time to enter a ticket.

‍

Tarush Aggarwal (14:40)
easier to calculate because you have the average time to enter tickets. We put also labels to say if it's good or bad for us to really improve the LLM. And then the last part is the number of tickets you can handle per month. So we are still scaling, but I can give you a first ROI, which I think is really interesting. So we decreased by 50% the time to...

‍

Anaïs Ghelfi (14:44)
We put also labels to say if it's good or bad for us to really improve the LLM. And then the last part is the number of tickets you can handle per month. So we are still scaling, but I can give you a first ROI, which I think is really interesting. So we decreased by 50% the time to handle a ticket from a customer care representative, which is a first step.

‍

Tarush Aggarwal (15:08)
to handle the tickets from the customer care representative, which is a first step because we have a lot of improvement to do. But I think that's a really good KPI to keep following. Yeah. That's incredible. Do you have plans? You know, this is a very fixed use case, which, you know, on customer success, on customer service, you can measure.

‍

Anaïs Ghelfi (15:13)
because we have a lot of improvement to do. But I think that's a really good KPI to keep following.

‍

Tarush Aggarwal (15:30)
As you think about the general adoption of this tool, are there any other use cases which are particularly interesting as you think about scaling this out? you can also go attribute an ROI on? Yeah, another way to calculate an ROI, so I'm not really answering your question but I'll explain my keynote thought, is coming back to the knowledge, right?

‍

Anaïs Ghelfi (15:44)
Yeah, another way to calculate an ROI. So I'm not really answering your question, but I'll explain my chain of thoughts is the when coming back to the knowledge, right? Before we were in a place where when you wanted to find information about the product and very specific question, you will like go on Notion, spend a lot of time and ask questions to someone directly on stack or open a ticket.

‍

Tarush Aggarwal (15:57)
Before, we were in a place where when you wanted to find information about the product, a very specific question, you will go on Notion, spend a lot of time and ask questions to someone directly on Stack or open a defect.

‍

Anaïs Ghelfi (16:11)
And KPI I'm tracking for the product knowledge adoption is the number of questions I got per week and the number of total users. Because I'm thinking, like, if someone is coming back, it means that the person has found value in the product, right? Otherwise, you come back. You come once and then you don't come back. And what I can tell you is the monthly average users we have is quite insane because it's, as of today, would be like 200, between 200 and 300 person.

‍

Tarush Aggarwal (16:12)
And the KPI I'm tracking for the product -to -lage adoption is the number of questions I got per week and the number of total Because I'm thinking, if someone is coming back, it means that the person has found value in the product, right? Otherwise, you come back. You come once and then you don't come back. And what I can tell you is the monthly average users we have is quite insane because as of today, would be between 200 and 300.

person, knowing that in a company, it's almost half of the company that is using the tool for product, especially product knowledge every month. So I say it's difficult to calculate the ROI, so the time we save, but at least we know that we are bringing value and another way to do it is NPS. Yeah.

‍

Anaïs Ghelfi (16:40)
knowing that in the company, so it's almost half of the company that is using the tool for product, specifically product knowledge every month. So I say it's difficult to calculate the airwaves or the time we save, but at least we know that we are bringing value and another way to do it is NPS.

‍

Tarush Aggarwal (17:00)
That makes a ton of sense. What is one challenge which you're currently facing, which is something which you are very invested in solving? So I'll say it's not the most, like I'll say, sexy subject, but I think when you work in the data platform team, sometimes you have to build foundations that nobody sees, but that's really important to do.

‍

Anaïs Ghelfi (17:09)
So I'll say it's not the most, I'll say, sexy subject, but I think when you work in the data platform team, sometimes you have to build foundations that nobody sees, but that's really important to do when it comes to stabilizing, making sure we scale and so on. And something I...

‍

Tarush Aggarwal (17:28)
when it comes to stabilizing, making sure it's scalable.

And something I really like to focus on and we start focusing on is really developing the data product part. Because for me, building a data product, you really showcase how strong is your platform, right? Because when you build that, you answer a business use case, you have ROI, but then you are leveraging all the capabilities the team is already building and where all the stack is on top. So I think something that I really like to foster is to make sure that we

‍

Anaïs Ghelfi (17:35)
to make sure we build more data products, so with a concrete use case. And also, I'd like to make the user more and more autonomous on our stack. So meaning you're coming tomorrow, you're like, okay, I'd like to automate this task. I'd like to use the products you've done on AI because I want, let's say I'm a sales, I want to customize my sales workflow. I'd like these persons to come, pick the components on self-service part and try test.

‍

Tarush Aggarwal (18:04)
more data products, so with a concrete use case. And also I like to make the user more and more autonomous on our stack. So meaning you're coming to more, you're like, okay, I'd like to automate this task, I'd like to use the product you've done on AI because I want, let's say I'm a sales, I want to customize my sales workflow. I like these persons to come.

pick the components on Self service path and try test and if it works, okay, we can discuss on scaling. So this is really something about a Self service platform to make people autonomous. That makes a ton of sense. And as you think about this project, you have, you know, sort of how long have you been at it and what would be the next metric as you think about?

‍

Anaïs Ghelfi (18:32)
And if it works, OK, we can discuss on scaling. So this is really something about a self-service platform to make people autonomous.

‍

Tarush Aggarwal (18:53)
adoption of this.

‍

Anaïs Ghelfi (18:56)
So looking at the product we built that is enabling everybody to use Gen AI where I told you we have a knowledge focus, which is on different departments in the organization, and another one which is on how do we build automation with that, with the customer care example I gave you.

‍

Tarush Aggarwal (18:56)
So looking at the product we built that is enabling everybody to use the AI, where I told you we have a knowledge focus, which is on different departments in the organization, and another one which is on how do we build automation with that, with the customer care example I you. I think a next step of that is really to see what are the other workflows we can simplify. And I can think of one. We know we in

‍

Anaïs Ghelfi (19:15)
I think the next step of that is really to see what are the other workflows we can simplify. And I can think of one. We know we are on multi-duet transactions. So we have a team that is focusing on everything related to payment delay, problems in financing and so on. So one of the next steps would be to say, okay, can we work with them and empower them as we did with the customer care?

‍

Tarush Aggarwal (19:25)
So we have a team that is focusing on everything related to payment delay, problems in financing and so on. So one of the next step would be to say, can we work with them and empower them as we did with the customer care to really make sure we reduce efficiency? And then it's really easy to find ROI behind that because you look at the number of requests handled, complexity of requests and stuff like that. So that will be another use case. And of course you can expand it.

‍

Anaïs Ghelfi (19:43)
to really make sure we reduce efficiency. And then it's really easy to find ROI behind that because you look at the number of requests handled, complexity of requests and stuff like that. So that will be another use case. And of course, you can expand it to many, many aspects of the processes.

‍

Tarush Aggarwal (19:55)
to many many aspects of the processes. That makes a ton of sense. Anais, thank you so much for being on the show and adding value to our listeners today. Thank you for having me. Of course.

‍

Anaïs Ghelfi (20:06)
Thank you for having me.