2.2: The data silo challenge: Wrest back control

As banks continue efforts to break down data silos, negotiating better terms with archive and surveillance tech suppliers can offer another way for them to cut costs and gain more control over their own data.

Growing regulatory requirements have dramatically driven up the cost of running a comprehensive surveillance function, both in terms of tech investment and the human resources required to filter through a ballooning number of alerts. Many banks have sought to plug gaps in their coverage by building point solutions onto legacy tech. This has exacerbated an existing data silo challenge, with the typical tier 1 bank now running countless voice-recording, trade surveillance and data-archiving systems that each format and time-stamp data differently as well as reference individual employees that are monitored in inconsistent ways. With many of these solutions, banks face an additional hefty cost, plus a long delay, if they need to extract data for their own analysis.

The result is a disparate mess that banks are rightly attempting to clean up by taking on the lengthy task of reducing the number of data silos they work with, according to Rob Houghton, founder and CTO of SOTERIA. While these efforts continue, however, banks can also gain more value from the data they generate, as well as trim costs from the function, by negotiating back control of that data from the myriad of tech vendors and applications that currently have custody of it, he argues.

“As banks measure their cost of compliance and decide how to manage that moving forward, it’s vital for them to understand the attributable costs for each silo they work with so they can plan for the future,” Houghton says.

Prohibitive export costs
The terms of many current framework agreements between banks and their archive and surveillance tech suppliers stipulate that data can only be extracted before the end of the contract if banks pay a charge that can stretch into hundreds of thousands of pounds for order management systems, and even higher for data-heavy voice recorders. Even after this, it can take months for them to access the data, which at this point is still in proprietary format. Changing the way systems record or format data can incur an extra charge. “If you have to do a discovery, the cost of that export process is ridiculous,” Houghton notes.

As contracts end, therefore, banks should consider negotiating hard for new terms that allow them to change the way data is formatted or to pull information out of applications – either continuously or when required – so that a second, bank-owned copy can be created and maintained, Houghton advises. Data in this copy can then be normalised and pooled with other normalised data sets so that anyone in the bank with analysis tools – from within compliance and beyond – is able to interrogate it as one. This is the key reason SOTERIA was designed to create an immutable ledger with integrated reporting and audit structures, he notes.

“While work to tackle data silos remains underway – and even if you take the strategic decision to maintain some silos –  being able to extract data at any point you like must be part of your contract with suppliers,” Houghton stresses.

Tilt the power balance
An advantage of this approach is that it helps shift the power with tech suppliers back in the favour of banks as suppliers realise they can be more easily replaced, argues Houghton. “Because you’re running all your reporting and analysis out of your own holistic copy of the data, you can chop and choose your supplier to be constantly working with best-in-practice, best-in-class firms,” he says. That should also help drive down costs.

Once banks have created a holistic data store that they own – albeit in copy form – they can also render it immutable, meaning nothing can be modified without leaving a clear audit trail, Houghton says.

The cost implications of maintaining copies also tend not to be as heavy as imagined, at least for less data-intensive applications. Although the data associated with voice and e-comms surveillance systems can be huge, “having two copies of an order management system or your metadata associated with every other activity you do isn’t a great deal of data, no matter how big a bank you are”, Houghton notes, adding: “Contrary to common belief, silos of data don’t cost that much money anymore because disk storage is relatively cheap.” 

Whatever approach banks take to the data silo challenge – whether fast-tracking the break-up of silos or opting for a more gradual reduction while negotiating better access to data – an essential first step is the creation of a golden source.  Once banks have control of all key metadata associated with silos, they can build a fully integrated solution that delivers an interpretation of regulations, the required policies related to individuals’ roles and the organisation’s stance towards its corporate responsibilities. 

Based on an international benchmarking survey collecting the views of industry leading experts from 15 of the largest financial institutions globally, the 2020 Surveillance Benchmark Report provides a unique insight into the maturity and development of surveillance functions over the last 12 months, as well as predictions for the future. Including in-depth commentary from regulators, practitioners, consultants and technology experts, it is the only report for professionals in the industry.

Lead sponsor


Partner sponsors



Eventus Systems logo 1


Researched and published by