What is our primary use case?
Kount is a better solution for us because they are really focused on maximizing sales. A lot of the other proposals that we saw were very focused on automation, machine learning, and driving down the chargeback rate. Of course, that's one goal, but it needs to be a balance between the two, between minimizing your chargeback and your risk but then also enabling as many sales as possible. If I wanted to minimize my chargebacks, I could just turn away and decline a lot of good sales. We need that balance. Kount did the best job marketing and highlighting that it was a solution where they offer AI and machine learning but it's within their tools that if you want to automate things, you can. We took the approach of automating something but we really thought that the human component was important for us to be able to manage the different levels of fraud.
Kount helped us position ourselves in a way where we could use humans. The humans could leverage the technology that Kount provided and are able to adapt and customize it to our needs. We could constantly change things and change the scoring. There's a robust offering of the different things that you can do with Kount whether you want a scoring model with different weights or you want to just set up a ruleset that automates things. The flexibility was what we really enjoyed. It grew over time with us because we started with them around about eight years ago now and so their tool also evolved immensely. They're very clear about their roadmap, they provide updates, and they're always striving to be better. Their focus on sales enablement was just as important as minimizing fraud and chargebacks. Whereas a lot of the other providers were just focused on minimizing fraud.
How has it helped my organization?
Kount enabled us to reduce our false positives by around 45% for fraudulent transactions, both false positives and false negatives. It helped us determine, make better decisions across the board, and protect our business and our integrity with our customers. We weren't insulting customers or turning good customers away.
We were also able to reduce our manual review rate exponentially. When we first started with them, our manual review rate was something crazy, like 55%, and Kount brought that down to 25%. From 25%, most recently, we were down to 4%. Over the course of those years, it was around 55%, 56% all the way down to 4% which was incredibly helpful for us. That's obviously a huge decrease in the workload.
The customizable rules are just as important because it overlaps with the user-defined field where you can make them very specific to your business. You can alter them and edit them, turn them on, turn them off, whenever you'd like. You can even get it down to almost a scheduling tool where you can say, "Okay, on Monday through Friday, run this rule. Friday to Sunday, run this other rule set." And so you can manage the different fraud trends that you're seeing that way. It helps you keep your manual overview rate as low as possible.
Another great feature is their ability to integrate with other third-party solutions like Ekata and Chargebacks911. Ekata is an ID verification and they all integrate together so you can just click on a link and all of your transaction data is automatically populated. Kount a very inclusive solution with a lot of different features.
What is most valuable?
The most valuable feature is its ability to create your own ruleset and edit it whenever you want. Their reporting functionality is very helpful. It's really robust.
There's a feature that's user-defined fields where it allows you the option to customize your needs. If I had a specific field that wasn't integrated initially from the standard data points that Kount requires, there might be different companies that have different data points in their order management systems that aren't universal. They're very specific to the business and Kount has user-defined fields where it's very easy for your internal development team to use the user-defined field and filter in whatever data point you want. You can get very granular into the specific needs that you need in the trends that you're seeing for fraud.
What needs improvement?
They could do a little bit better with chargeback management. There are other solutions out there that I've heard about, like Accertify that have a better chargeback platform where they're integrated more with the banks or in terms of how the workflow is and how you can respond to chargebacks. From my experience, Kount doesn't really have that. They had a way to assign reason codes and descriptions to the chargebacks so you can run reporting. You can flag transactions that get chargebacks on them so that you can then report your chargeback trends and create rulesets around them but there wasn't much for actual chargeback management. That was a whole separate process that we had to have in place.
They should have more dashboards for data visualization. They have some capabilities there, but you really have to be a pro user. You have to be an expert user to really use it at the highest level. Otherwise, you're just exporting the data and creating it in Excel and using those visuals. It would be nice if they had some out-of-the-box dashboards where could you sign in, go to your specific dashboard and it just tell you everything in an intuitive way. That's a key area that they could work on a little bit more. The technology is there, it's a great solution, but as far as communicating the performance of the fraud department, they could make the dashboards more intuitive and insightful.
What do I think about the stability of the solution?
Kount does a good job of communicating when they're going to be performing maintenance and that's usually during non-peak times. Other than that, we did have some challenges at the beginning where their reporting functionality was really slow. And there's a latency time. I'm on the East coast and the reporting happens on the Pacific Time, so it's three hours earlier, and there's also a random latency time where sometimes we have to wait. Our team had to wait four hours before we could be confident in the numbers and so that was a little bit challenging. Directionally, the reporting was always correct but it wasn't exact. That would be one unfortunate thing that they could focus on.
As far as the reliability and stability of the tool itself, I can't recall a time that it ever went down. It was always able to handle our peak volume times during key retail sales holidays like Black Friday and Cyber Monday. It was able to take on an exponential increase in transaction volume. It was the reporting functionality that was a little bit lagging at times. They did do an overhaul where they tried to improve that performance and it did help. It still wasn't 100% exactly what I would have loved to have seen, but it was nice to have a little bit of an improvement to begin with.
What do I think about the scalability of the solution?
It definitely is scalable. Especially as a platform, we would implement new clients and onboard new clients and we never had a problem scaling. There was even an option that was very straightforward for us to either have some clients on our platform. We did their fraud management for them and other clients but they wanted to do their fraud management in-house. And so some clients would roll up under a parent account for Kount. Kount made it very easy to break other clients into their own individual instances. It's very flexible, very accommodating, and very easy to scale.
How was the initial setup?
The initial setup was very straightforward. The technical people provided the appropriate documentation. Even after the initial setup, throughout our relationship with them, they would always make sure that they communicated when certificates needed to be updated or if we weren't on the certificate model and we moved to a better API connection. If anything needed to be updated, they were always really good at communicating that. The initial setup seemed to be a really straightforward process. It was a little bit unique for our use case just because we were a platform provider. At the time, Kount was used to working with individual retailers. We were one partner of theirs that was bringing multiple customer accounts to them through us. It was a little bit of a unique use case. The implementation, it didn't go beyond scope. It was right on time, on budget. They're really customer-focused. They have wonderful account managers there and they're a pleasure to work with.
Our strategy was to scope out the projects and get an understanding from Kount exactly what was required of us. Our development team worked it into their specific sprints, but it only took about two or three sprints. We have two-week sprints so I don't think it took any longer than three sprints for us to integrate, test, and be fully live in production. They do have a nice sandbox environment. The great thing about Kount is that you can set your ruleset in production as test rules. You don't have to do everything in testing and then replicate it over to production. After the initial implementation, when you make changes, you can do everything in production, as long as the initial integration is set up.
What was our ROI?
We have 100% seen ROI. It's paid dividends over what we've invested just because it helped us quickly identify our areas of fraud and where we were being the most impacted and losing money. They helped us put parameters in place that stopped the flood gates. It stopped us from losing money. We started doing business smarter and saving good customers, boosting our sales, and minimizing the fraud at the same time. Kount is absolutely worth the investment.
What's my experience with pricing, setup cost, and licensing?
Pricing is very realistic. We were a platform provider. When we started, we had a normal account setup that a normal retailer would have. As we grew over time, we actually expanded to their partner portfolio, meaning that we were bringing them a lot of business as a platform provider ourselves. So we benefited from the increase in transaction volumes across so many clients that it had helped us cost-wise. We had a higher volume so our transaction fees were pretty low.
Overall, the pricing is very reasonable especially for the value that they bring. They do a wonderful roadshow every year and they come to your location or major citiy. They also do a great data conference.
What other advice do I have?
My advice would be to make sure when you're going through the proposal process to outline certain use cases that are specific to your business. I'd also suggest trying to speak with other customers that Kount has on their platform that are in the same industry as you.
Try to connect and reach out to them, because Kount does a really good job but they work with so many different industries that you want to make sure that they're specific to your business and that your pain points will be addressed.
They do a good job covering all of their bases. It's really just about if there are specific pain points that are particular to your business or your industry, make sure that you get some very valid examples to show how they're solving them.
I would rate Kount a nine out of ten.