The fact that, to this day, not a single foreign account of these Wirecard third-party partners is available, and that after two years of investigations, not a single operational account has been obtained from the public prosecutor's office by the largest Wirecard third-party partner, the company Senjo, is no longer comprehensible from any point of view.  

Bellenhaus Data Warehouse with over 2 billion transaction data points per year from third-party business

Further evidence that the TPA business actually existed is provided by the following:
Email communication and statements made by witness Gavin Jackson to Gleiss Lutz demonstrate that between 2016 and 2019, Bellenhaus had over 2 billion genuine transaction data points from its third-party business backed up annually in a data warehouse ("Hadoop") and aggregated there for analysis. The company Syncrasy, initially operating under the name TeleDyna-mix, was commissioned to do this. Witness Jackson stated the following to Gleiss Lutz (BB XXX-1-8, 3681, 3682):  

“I was personally involved in the ‘Hadoop database’ project when it started in 2016. Mr. Bellenhaus approached us – but I can no longer recall which Wirecard company he was acting on behalf of. In a conference call, Mr. Bellenhaus explained that Wirecard needed a platform (data warehouse) to process large amounts of data so that it could then be analyzed quickly and easily.”  

What did we do? We looked at each of the files, line by line, and  

"They aggregated the data in a data warehouse. Mr. Bellenhaus and his team were then able to analyze the figures and compile statistics." (Emphasis added)

Mr. Bellenhaus also confirmed this to KPMG. For example, in a meeting on December 10, 2019, he stated that since 2016, a quarterly review of the transactions provided by the TPAs ​​had been conducted, meaning the data had been reconciled with the statements. The minutes prepared by KPMG state specifically (BB XXX VIII, p. 22):  

“In response to the question regarding the design of the internal control system with respect to accounting in the context of the business relationship with Al Alam, Mr. Bellenhaus explained, in essence, that controls had been introduced successively between 2016 and 2018. For this purpose, CSME had developed analyses based on the HADOOP transaction database using the TABLEAU reporting tool.”  

The completeness of transactions is verified quarterly by reconciling the transaction volume (using the "settlement information file" generated on the e-payment platform) with the quarterly statements from the respective partners (e.g., Al Alam). A materiality threshold of 5% has been set for this purpose. Deviations arising for various reasons regularly amount to 1%–2%.  

According to witness Jackson, the data was delivered anonymously, meaning the merchant names were coded. However, they were able to initiate chargebacks.  

in the transaction data (BB XXX-1-8, p. 3682):

“We were able to identify chargebacks in this transaction data. We could see whether payments were authorized, declined, reversed, or refunded. The chargebacks varied. The chargeback rates were between 4% and 8%, as far as I remember. I was told that there would be problems if the chargeback rates from individual merchants were too high. In these cases, Mastercard and Visa would reject the payments.”  

The fact that chargebacks were evident from the data indicates that it was real transaction data. Bellenhaus stated to witness Gavin Jackson that this data came from high-risk business that had to be outsourced. Wirecard received large amounts of transaction data from third-party partners via various systems (BB XXX-1-8, p. 3681). Accordingly, the underlying contract also contains a clause stating that the transaction data is provided by the third-party partners (BB XXX-1-6, p. 2579).  

“The establishment of a platform for the facilitation of the analysis of credit card transactions provided by 3rd parties in text format, to provide reasonable historical business trends for audit and business planning purposes.”  

According to Jackson, the volume of transaction data doubled between 2016 and 2018. Witness Jackson stated the following to Gleiss Lutz on October 8, 2020 (BB XXX-1-8, p. 3682):  

“We were told that the transactions were high-risk business that needed to be outsourced. This isn't unusual. We were only tasked with aggregating the data, not with verifying its authenticity. To me, it felt like genuine business. Oliver Bellenhaus dedicated a considerable amount of time to the project; he thoroughly examined the issues.”  

I didn't know that the data pertained to Wirecard's business, which accounts for 50% of its revenue. We couldn't deduce the revenue from the data. We looked at 2 billion transactions per year, contained in approximately 200,000 electronic files. (Emphasis added)  

“I seem to recall that the volume nearly doubled in 2018 compared to 2016.” (Emphasis added)

The communication also reveals that Bellenhaus specifically attributed the data to the TPA partners. For example, on January 29, 2018, he wrote to Jackson:

“Please discard all actual Senjo data”  

After Jackson stated that no Senjo data was available for 2017, David Yasmineh replied that it would be provided (EA III-1-I, p. 499).  

The first email correspondence dates back to early February 2016. The request apparently involved the aggregation of data from 2014, 2015, and 2016 (BB XXX-1-6, pp. 2469, 2571). The data was sorted by merchant, date, and transaction status. Jackson wrote on February 8, 2016:

“I am enclosing a spreadsheet of the Date Analyzed by Merchant, Date with the columns of The Totals: Approved, Charge-back, Declined, Void, Refund”  

Bellenhaus analyzed this data and used it for representational purposes, such as presentations about the TPA business (see, e.g., BB XXX-1-6, pp. 2603 ff.).

Apparently, data from each month was also available, as witness Jackson stated that Bellenhaus checked the data to see if any was missing (BB XXX-1-8, pp. 3681, 3682). It can be assumed that a file was created for each month of the year. Jackson wrote to Bellenhaus on October 2, 2019 (BB XXX-1-7, p. 2883):

“We are close to completing the project. The system ingests all 12 files from 2018.”  

The email correspondence between Bellenhaus and Jackson also proves that data from 2018 was aggregated to analyze buying habits well before the KPMG audit. For 2018 alone, there were 300,719,827 valid transaction records and an additional 79,750,302 transactions where the credit card holder could not be identified (BB XXX-1-7, p. 2883).

These must therefore have been genuine transactions from typical TPA business, because an analysis of customer purchasing behavior and its use for presentation purposes is only possible with authentic data.  

Bellenhaus did not disclose the source of this data during his interrogation. For 2016, he claimed to have received it from Marsalek. However, he also stated that he was given a sales figure and then the data intended to represent that sales figure (EA III-1-II, p. 477).

In 2019, the data was allegedly generated by a data generator (EA III-1-II, p. 477):

"This is the situation again in 2019; the data for this database came from the data generator I already mentioned or from the Elastic Engine."  

The idea that such a large amount of transaction data could have been artificially generated is completely implausible. Bellenhaus's description of the data generator's function alone demonstrates that it is impossible to artificially generate such a volume of data. The data generator was allegedly used for the 200 million data records for the KPMG special audit. For this purpose, a read file in the form of a text file containing the transaction data from an invoice was supposed to be uploaded to the data generator, which would then generate the transactions based on this data (EA III-1-II, p. 474). Finally, Bellenhaus reveals that even generating 200 million data records was so complex for KPMG that the data generator had to be upgraded to a higher level (EA III-1-II, p. 488).  

t is obvious that the story about the data generator is a lie, intended to conceal the existence of genuine transaction data originating from real third-party business.

The fact that this data must be authentic is also evidenced by the minutes of the quarterly meetings with the third-party partners, written by Bellenhaus. Virtually all of these minutes contain statements regarding...  

The data delivery can be found in the logs. Many logs also include screenshots from the Tableau software showing the number of transactions, merchant IDs, volume, month, and chargebacks (e.g., Chargeback Summary JPY Q1 Al Alam BB XXVII-1, p. 157). Furthermore, graphs illustrating the development of transactions based on this data are included (e.g., Performance Senjo Q2 2016 BB XXVII-1, pp. 106, 107). The logs were also provided to KPMG and EY. It is inconceivable that this is "just any" or even artificially generated data unrelated to Wirecard's third-party business, and that this could have remained hidden from the auditors.  

Bellenhaus deleted the transaction data, which served as evidence for the existence of the third-party business, before his first interrogation in order to prevent the investigating authorities from accessing it. The following findings can be found in the so-called Steinhoff Report (pp. 84 ff.):

“The external service provider named by Mr. Oliver Bellenhaus is Syncerasy. In February 2020, KPMG attended a meeting with the CEO of Syncerasy, Mr. Gavin Jackson, during which a presentation on the functionality of the software was given. …  

The data was then processed via a virtual desktop and subsequently loaded into a data warehouse. However, access to transaction data and the data warehouse is now lost. The server was shut down by Mr. Oliver Bellenhaus's team after the escrow account fraud came to light. (Emphasis added)