Accarda and Exasol

Streamlining processes in order to avoid losses in inventory

(Un)intentional errors at the checkout when collecting payments or when booking returns, reversals or special discounts can lead to discrepancies in inventory and losses in any retail business. In Switzerland, inventory discrepancies currently account for as much as one percent of total retail sales, and the trend is increasing. Such discrepancies can quickly translate into financial losses as well as an overall loss of confidence within any company. Responding to this problem, the financial service provider Accarda, a specialist for payment solutions and customer loyalty in the retail sector, has worked with Manor AG to develop the Accarda Loss Prevention System (ALPS). ALPS tracks down improper or incorrect transactions, and can provide valuable information in order to detect fraudulent acts.


Challenges

Accarda needed a scalable solution that they could use to run any type of analysis at the individual transaction level as well as analyze data on the fly.

Solutions

The solution runs on two clusters, each with five nodes and has integrated seamlessly into Accarda’s existing infrastructure.

Benefits

Accarda can now speed up loading and analysis times even more.

With Exasol, we have managed to accelerate analytic query response times quite considerably. There’s a remarkable agility not only in the development of the product, but also in Exasol, something that we have come to notice a lot.

Stefan Schurgast Business Development & Innovation Manager at Accarda AG
0

Clusters

0

Nodes in each cluster

Background

In order to uncover improprieties, investigate discrepancies in processes or identify any conspicuous trends, Accarda needed a scalable solution that they could use to run any type of analysis at the individual transaction level as well as analyze data on the fly. Loss prevention systems help companies to reduce losses caused by erroneous checkout transactions, such as manual price changes, unauthorized receipt cancellations, void receipts or return fraud. Considerable cash savings can be achieved if improper action is detected early on and if countermeasures are taken in time. This requires analytics and tools able to analyze huge data volumes in real time without first having to spend time aggregating data. With this in mind, Stefan Schurgast, Business Development & Innovation Manager at Accarda AG, and his team had been searching for a new suitable, column-oriented database.

  • Industry Financial Services
  • Application BI Acceleration
Case study background details

The journey

You can use our analytics database with our private cloud, any public cloud, on premises – or as part of a hybrid deployment.

The challenge

Their extensive search resulted in a shortlist made up of two service providers. Schurgast sums up the requirements that the new solution had to meet: “Every day, a vast number of checkout transactions are performed in retail. As a result, our customers accumulate large sets of data that have to be loaded easily into a database. Another important factor was that the solution we would choose had to offer us the ability to run statistical analytics, reporting and ad hoc analyses.“ Finally, it was Exasol and its solution that proved to offer the widest range of support for our needs.

The solution

Agile and compatible

The Exasol-based Accarda Loss Prevention System has been used in production since the beginning of 2014. Asked about the criteria that prompted Accarda to opt for the Loading...Loading...in-memory database, Schurgast explains: “We were not only impressed by the high loading and analysis speeds, but also by how compatible the database is with existing systems. We like both the efficient use of storage and the price/ performance ratio. As it is linearly scalable, the solution grows in line with every new customer that uses ALPS. There’s a remarkable agility not only in the development of the product, but also in Exasol, something that we have come to notice a lot.”

The new solution runs on two clusters, each with five nodes. The scalable database has integrated seamlessly into Accarda’s existing infrastructure and the company uses it largely to analyze Loading...big data volumes.

The benefits for Accarda

Accelerated data analytics thanks to Exasol

The primary advantage for Accarda is that data aggregation has become superfluous and data does not have to be materialized. Accarda can now speed up loading and analysis times even more. “With Exasol, we have managed to accelerate analytic query response times quite considerably, and our trading partners have benefited as a result. Also, we can analyze down at the individual transaction level, something that we will continue to benefit from in the future.”

How ALPS works

A quick summary of how ALPS works: First, the customer delivers checkout data to Accarda. The payment solution specialist then processes the data and analyzes it to detect any transaction interruptions, key figures, irregularities or exceeded thresholds. Any suspicious checkout transaction or event identified by ALPS is then automatically shown in a report and the departments concerned are alerted accordingly.

“Customers do not only receive specific daily target reports from Accarda; they are also enabled to conduct in-depth ad hoc analyses themselves, so they can initiate formal investigations, put in place sanctions and take measures to prevent theft or similar offenses,” says Schurgast. When asked about Exasol’s technical advantages over other systems, Schurgast is quick to respond: “The solution’s compatibility with Oracle is tremendously helpful; Exasol uses standard SQL and runs on standard hardware, and we can continue to use our existing ETL tools.”

The future

Therefore, Accarda plans to implement more Exasol-based analytic systems over time. In this regard, Schurgast emphasizes the great architecture approach that Exasol has followed when developing its solution. “In my search for a column-oriented database, Exasol caught my eye. Yet, what convinced me in particular was the solution’s four technical building blocks – clustering, Loading...MPP, column-oriented storage and smart compression algorithms – as well as the Loading...in-memory component.”

Start your Journey

Get in touch today

We’d love to hear from you if you’ve got a burning question for one of our experts – or if you simply want to talk through how to take your project forward with us, or one of our partners.

Contact Us