What is our primary use case?
We have been using it for our risk management portfolio. We are a lending institution. We give credit to small and medium enterprises. We've been using it mainly for client segmentation and the probability of delinquency in the loans that we get.
I am using the latest version.
How has it helped my organization?
Due to the predictions that we have been able to do because of the use of Darwin, we have decreased our delinquency index. We were almost at nine percent by July. After using Darwin, we have reduced the delinquency index to five percent. Darwin is important for this metric.
We calculate the potential loss of clients. When some of our clients are going to go with competitors, they will ask for money with another bank or institution. In July, we had this number at 19 percent of our clients. We saw that they were likely to go with some of our competitors. Now, after using Darwin to narrow and segment more of our clients, we have been contacting them in a more specific way and targeting them more specifically. So, our clients loss index has gone down to 10 percent. This is also important.
I like the solution’s automatic assessment of the quality of data sets for uncovering issues like missing data points, low statistical variance, or incorrect data types because of Darwin's process capacity of filling in the data for us or even giving it double or triple checks. However, I still believe that the data cleaning process should be cleaner and better before it arrives to Darwin. We still have a lot of processes to do before we put the data sets on Darwin.
We have a process which gives us the credit score for all the credit companies in Mexico. We have to clean the data first through Darwin, so that we can use it in our modeling. Darwin has been very helpful on cleaning this data with simple queries.
We use it more for production of models. We have not had any trouble whatsoever with the data repository connection. It's consuming the data sets correctly, which is good.
It does affect the decision-making of our risk committee using the new models. They adjust the data accordingly, then this committee reviews everything each week so we can be fast enough to compensate for any new approach or findings that we have through Darwin. It is constantly optimizing the model, so once a week has worked for us.
Darwin allows us to make better decisions and more quickly.
I believe that Darwin is helping us to evaluate all our processes in a much faster way. That is why it affects our productivity for the good. It solves a lot of the time consumption. This really makes a difference for us: the time consumption. The processes, evaluation of the processes, and results may be the same, but the time it takes us to process it through Darwin rather than on traditional tools or other tools is so radically different. It is changing the way we make decisions because we can make decisions faster, and maybe, that's the most valuable thing about Darwin.
What is most valuable?
The model processing is valuable. However, what I find most valuable of all is the time. With the team, we could have maybe reached these numbers, but it would have taken double or triple the time to reach these numbers. So, with the Darwin tool, we are able to test our models constantly. We can go with the optimal way in minutes. That has been a game changer for us.
The tool is very powerful and has many benefits. The time reduction in the modeling testing is the most valuable thing at the moment. The time that Darwin saves for us to be constantly testing the model has been a game changer for us. We could have reached these numbers maybe with qualitative analysis, but it would definitely be with more time. With Darwin, we are reaching these numbers very quickly.
The solution tracks the health of models. We're also looking into the possibility of having alerts. So, when Darwin can find an optimal model better than what we currently use, it lets us know.
What needs improvement?
The challenge is very big toward making models operational or to industrialize them. E.g., what we want to do is to make unique credit models for each customer. So, we are preparing the types of customers who we can try new credit models on Darwin. But, I see this still very challenging to be able to get the data sets so Darwin can work. At this point, we are working with it to get the data sets ready for Darwin. However, once they are in Darwin, I believe we will not have any problems and will have very good results, just as we have had for the risk portfolio management. We are trying to aim it for a more specific group of clients to target them more specifically.
Right now, we have been using Darwin with clients that we don't want. That's how we have been reducing our delinquency index. Darwin is helping us identify clients that we need to close a relationship with, but we need it now to tell us the clients where we should be aiming to give them new products, new opportunities, or go to the market and reach new clients.
The dashboards and displaying of the data needs improvement. Currently, only IT and business intelligence people are using the results we get from Darwin, but less sophisticated areas in technology could also benefit from it if we had more user-friendly dashboards. People get scared, and they think, "We will need to run something in Python," which is not the case. We could use more user-friendly dashboards so everyone could use them. However, they have already let me know that Darwin is already working on the dashboard implementation so our commercial areas can have access to the data in a more user-friendly way. This is great because it is a very important area of opportunity.
We want to be able to test different updates. E.g., we've been waiting for the user-friendly dashboards since August. We really want to start working with that but don't know when it will be released. The people at SparkCognition told me that as soon as they were ready that they would contact us, so we could have a workshop for this. However, they haven't contacted us for this yet.
For how long have I used the solution?
I have been using the solution for six months.
What do I think about the stability of the solution?
We find it stable. I haven't heard of any technical problems. The availability is more than 99 percent.
What do I think about the scalability of the solution?
The solution’s ability to capture complex relationships over time and the resulting accuracy of its predictions is excellent. We don't have restrictions on the amount of data or processes that we can run at the same time. The opportunities are limitless. In fact, what I will be working on is getting all new data sets with all new correlations that we want to try on the modeling, as we are seeing that the tool is even more powerful and we can get more benefits from it than the ones we are currently getting. So, we are getting ready many other data sets to test them.
Currently, only the business intelligence team and risk management team are using Darwin. But, we want to see the dashboards in the near future so we can have the implementation for the commercial area too. Because right now, the final user is not reaching the data. So, the next step is to let the final user reach the data and make better decisions with it.
We have five licenses now: Three of them are for the business intelligence unit and two of them are for the risk management unit.
How are customer service and technical support?
Whatever problems that we have had, their team has given us super fast responses and are always willing to help us with our questions.
Which solution did I use previously and why did I switch?
We did not use another data intelligence solution before Darwin. We used regular traditional statistical tools, like Data Studio and R, but not something as powerful as Darwin.
How was the initial setup?
The initial setup was straightforward. We had no problems whatsoever. They were very clear on the instructions and there were minimal technical requirements to install.
It took us less than two hours to deploy Darwin. When we deployed the tool, we only used one person from IT operations to install everything.
What about the implementation team?
We had two workshops with the business intelligence team and technology team so the setup could be done prior to the workshop, then the workshop could be done with different models, etc.
We are actually aiming to have a new workshop.
What was our ROI?
In just six months, we calculated six million pesos that we have prevented in revenue from going away with another customer because of this solution. Thanks to Darwin, we didn't lose those six million pesos.
Darwin has increased efficiency and productivity for our company. With our risk management team, there were models that took them more than three days to process each, only to see the outcome. Now, it takes minutes for Darwin to process the current model. So, we can have it in minutes. We don't have to wait three days for all the models to be tested, then make a decision.
What's my experience with pricing, setup cost, and licensing?
The license cost is not cheap, especially not for markets like Mexico. But sometimes, you do have to make these leap of faith for some tools to see if they can get you the disruption that you are aiming for. The investment has paid off for us very well.
Which other solutions did I evaluate?
We went with Darwin for its potential of disrupting the market with a new type of technology that could allow us to give unique customer journey experiences, which is what we were aiming for.
We looked at Oracle and others who had their business intelligence solutions along with other things. However, we didn't feel that they were disruptive enough, especially not for our markets. We really do want to do things differently. We made it clear that we wanted to go where no one else was going. SparkCognition presented a very good tool to do that, which is why we wanted to invest in their licenses and start seeing what we can take out of them.
What other advice do I have?
Do not be intimidated by the apparent complexity of it because it is more user-friendly than you think. It makes AI easy. Start testing it because it's very trial and error. I really do believe people need to have this type of mentality to start using tools like Darwin. Don't be afraid of retesting it.
We are using the automated AI model building because we want the AI model to be unique for each customer. We are getting all the data ready so it can be integrated into the modeling. We want to give each client a unique credit model to be automated through the AI. We don't have this currently. We are working on it. Right now, we don't have this in production, but are working on it so we can get there.
I would rate them a nine (out of 10). I wouldn't put them as a 10 because there are still a lot of things for them to keep trying. However, so far, there are a lot of benefits that we could be taking out of it, but that is part of the learning process.
Which deployment model are you using for this solution?
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?