By Alvin Tan, principal consultant at Capco
Data sourcing and cleansing are frequently cited as being among the most critical, yet most time-consuming, aspects of data science. Enhanced data management not only reduces the burden of data sourcing and preparation, but also improves data quality and serves to foster greater trust in the insights that are delivered via data science.
Robust data management capabilities ensure that less time is spent wrangling data into an analytics model and more devoted to the actual modelling and identification of actionable business insights. Organisations that build analytics data pipelines upon solid data management foundations can extract greater business value from data science.
This delivers not only competitive advantage through newly-identified insights, but also a comparative advantage via a virtuous circle of data culture improvements.
Lies, damn lies and statistics
Data science is only effective if it ultimately delivers positive value: the analytics must serve a clear business purpose. ‘Garbage out’ – incorrect, misleading, meaning lessor otherwise unusable data science output – leads to sub-par strategies and misguided business decisions at best, and financial and reputational damage at worst.
This diminishes the business value of the results at hand and moreover erodes the faith that decision makers might place in future results. Trust is key. No matter how powerful, accurate or statistically sound the results, a data science capability must itself be trusted if decision makers are to transform those results into business strategies. Establishing, retaining and nurturing such trust requires business outcomes to be consistently aligned with expectations.
Trust not only requires the adoption of sound scientific methodologies, but also a cost-effective mechanism of ensuring data issues are flagged, managed and resolved. This can be boiled down into two key data management requirements for analytics processes: understanding and obtaining the right data; and fixing the data that is generated.
- Understanding and obtaining the right data
In the case of model-led analytics (for example, machine learning) data is input into an existing analytical model to ascertain its accuracy and viability. In this paradigm, the data scientist must first understand the semantics of the data to be sourced, so that the conceptual and contextual specifics of the required data can be identified. The data scientist must then determine from where to source the specified data. That requires an understanding of data provenance if the data is to be sourced appropriately.
Comparably, an understanding of what data to source, and where to source it from, is also necessary to ensure outcomes of data-led analytics (such as data mining) have strong foundations. Therefore, an understanding of data semantics and data provenance are critical to ensure that any analytics draw upon the right data:
• The data required must be properly and unambiguously defined. This involves identifying and establishing a shared understanding with potential data providers as to what is required. If the data scientist wants ‘customer name’, for example, then an agreement must be made with the provider as to whether ‘name of account holder’ means the same thing semantically. In this example, there are many hidden nuances: does customer name include prospective or former customers? Does name of account holder cover mortgages, or current accounts, or both? Arriving at a mutual understanding is no simple task without a commonly agreed understanding of the definition, taxonomy, and ontology of the data.
• The data that is obtained must be representative of the population. An unrepresentative sample, for example where data obtained only represents specific subsets of the required population biases analytics outputs, should be avoided. As an example, if retail banking customer names are required, then it is important to ensure that the data is sourced from a provider that aggregates customers for all retail banking products, and not just, say, mortgages.Satisfactorily resolving this sourcing challenge requires not only an accurate semantic articulation of the required data, but also an understanding of where this data can be reliably obtained.
- Fixing the obtained data
Once sourced, data may still contain data quality issues that must be properly understood and resolved prior to any analytics. Resolving and correcting for data quality issues is a data cleansing process that constitutes a key element of analytics preparation.
Poor quality data inputs can manifest in a variety of ways:
- Data may contain gaps, which if not corrected at source, accurately input, or omitted entirely, will result in abiased output;
- Similarly, data may contain duplicate elements, which if not omitted will also lead to biases;
- Data may not conform to an expected format, which if not corrected may break the analytics model;
- Data may contain errors, which reduce the accuracy of the results;
- Data may be out of date, hence the relationships inferred may no longer be applicable;
- Data may not be sufficiently granular, or sample size may likewise be insufficient; both scenarios weaken explanatory power and the significance of outcomes.
‘Garbage in’ – incorrectly defined, inaccurate, incomplete or otherwise poor quality data entered into an analytics process – is a primary limiting factor on the usefulness and reliability of analytics results.
Managing the inputs
To avoid misleading data science outcomes that might drive bad business decisions, while also minimising the marginal cost of implementing such remedies, organisations must implement an effective data management capability, one that delivers the scale economies required to ensure additional data science projects are cost-effective.
A data management capability provides a set of centralised, scalable services for describing what data means; for understanding and recording where the data comes from; for maintaining good quality data; and for ensuring the roles and responsibilities for data management are effectively discharged.
- Semantics: data is given commonly agreed and understood definitions, is placed in a commonly known taxonomy and ontology so it can be categorised accordingly, and semantic relationships between data are clear;
- Provenance: the sources of data, and the paths to where it is consumed, are identified and documented;
- Quality: various quality dimensions such as completeness, conformity, consistency, validity, accuracy and timeliness of data are measured and published/reported on a regular basis;
- Governance: the decision-making bodies, policies, processes, accountabilities and responsibilities by which effective data management is defined, monitored, and enforced.
Without a vision for streamlining how these requirements are met, an organisation’s data science efforts can all too swiftly devolve into a web of hit-and-miss, fact-finding engagements between analytics projects and potential providers -with each project independently trying to surface the right data from the right sources. Analytics projects may even start ‘sourcing’ data from other analytics projects which becomes ingrained into ways of working, reinforcing bad habits in cultural norms that prevent development of a mature data-driven organisation.
Conversely, a centralised data management capability will provide a hub of data services and expertise that allows all processes –whether analytics or not – to outsource their data management requirements effectively and help foster a strong data culture.
There are several benefits here. Firstly, the data semantic (definition, taxonomy, ontology, and modeling) and data provenance (lineage and trusted sources) services offered not only frees up valuable time and effort, allowing data scientists to focus on the actual analytics, but also ensures more reliable and explicable analytics results.
Secondly, the hub serves as a governing body for all data management within the organisation, ensuring that the outcomes are available across all processes. This allows for incremental gains, with the knowledge (semantics, provenance, and quality) built for one project adding to the existing organisational body of knowledge in an accretive fashion.
Thirdly, a centralised data management capability allows analytics processes and models to be defined within a globally accepted semantic model. This allows analytics results to be defined and communicated in a common business language, which in turn enables better interpretation and understanding of results across different decision makers.
Improving data culture
Regardless of the trustworthiness of analytics results, decision makers do not habitually act on these insights. This is particularly the case with data mining insights, which are often produced in financial services with little business sponsorship and poorly defined or planned business implementation.
What is often missing is not just trust, but also a willingness among decision makers to put insights into practice. This reflects an inherent preference to stick with subjective, opinion-based decision making. This risk aversity towards relying on data science outcomes can be countered by having decision makers actively drive the data science process,and so be invested and interested in the outcomes.
Data governance is a key service that ensures the effective discharge of roles and responsibilities in relation to the management of data. Crucially, data owners and stewards must be identified and also engaged in the governance and management of data. These data owners are typically the same decision makers to whom analytics projects provide insights. In this way, the effective implementation of a data management service, helps to drive cultural improvements by ensuring decision makers actively participate in the governance of the organisation’s data.
Effective data management
A data management capability helps foster a data culture that places decision makers, rather than data scientists and data managers, at the forefront of data-driven decision making.It requires data owners to be involved in the governance of the data from which they draw their insights.
Trust is built by ensuring that business outcomes are consistently in line with expectations. This requires expectations to be properly set, which in turn requires the semantics, provenance, and quality of data inputs to the analytics be defined and known – ‘good’ inputs.
While very time-consuming and resource intensive when applied to each data project in a silo, outsourcing these data management requirements to a centralised data management function means economies of scale are achievable. Successful data management is the foundational layer for good data science and data-driven decision making.
How to Build an AI Strategy that Works
By Michael Chalmers, MD EMEA at Contino
Six steps to boosting digital transformation through AI
In the age of artificial intelligence, the way we interact with brands and go about our work and daily lives has changed. No longer blithe buzzwords, AI tools and algorithms are solving real business problems, streamlining operations, boosting productivity, improving customer experience, and creating opportunities for advantage in a competitive marketplace.
However, many businesses struggle to unlock the full benefits that come with its adoption across the whole organisation. Making the most of AI requires a strategic focus, alignment with the specific operating model of the business, and a plan to implement it in a way that delivers real value.
Not all AI strategies are equal. To be successful, businesses need to set out how the technology will achieve objectives and identify the specific assets and case uses that will set them apart from competitors. The process of creating and delivering a successful AI strategy includes the following six essential elements that will help to bake in business success.
- Start with your vision and objective
One slip-up companies often make when developing an AI strategy is a failure to match the vision to the execution. Almost inevitably, this results in disjointed and complicated AI programmes that can take years to consolidate. Choosing an AI solution based on defined business objectives established at the start of a project reduces the risk of delay and failure.
As with any project or initiative, it’s crucial to align your corporate strategy with measurable goals and objectives to guide your AI deployment. Once a strategy is set and proven, its much quicker and easier to roll it out across divisions and product teams, maximising its benefits.
- Build a multi-disciplinary team
AI is not an island. Multi-disciplinary teams are best placed to assess how the AI strategy can optimally serve their individual needs. Insights and inputs from web design, R&D and engineering will together ensure your plan hits objectives for key internal stakeholders.
It’s also important to recognise that with the best will and effort, the strategy might not be the perfect one first time around. Being prepared to iterate and flex the approach is a significant success factor. By fostering a culture of experimentation, your team will locate the right AI assets to form your unique competitive edge.
- Be selective about the problems you fix first
Selecting ‘lighthouse’ projects based on their overall goals and importance, size, likely duration, and data quality allow you to demonstrate the tangible benefits in a relatively short space of time. Not all problems can be fixed by AI, of course. But by identifying and addressing issues quickly and effectively, you can create beacons of AI capability that inspire others across the organisation.
Lighthouse projects should aim to be delivered in under eight weeks, instead of eight months. They will provide an immediate and tangible benefit for the business and your customers to be replicated elsewhere. These small wins sow the seeds of transformation that swell from the ground up, empowering small teams to grow in competency, autonomy and relatedness.
- Put the customer first, and measure accordingly
Customer-centricity is one of the most popular topics among today’s business leaders. Traditionally, businesses were much more product-centric than customer-centric. Somebody built products and then customers were found. Now, the customer is, and should be, at the heart of everything businesses do.
By taking a customer-centric approach, you will find that business drivers determine many technology decisions. When creating your AI strategy, create customer centric KPIs that align with the overall corporate objectives and continually measure product execution backwards through the value chain.
- Share skills and expertise at scale through an ‘AI community of practice’
The journey to business-wide AI adoption is iterative and continuous. Upon successful completion of a product, the team should evolve into what’s known as an ‘AI community of practice’, which will foster AI innovation and upskill future AI teams.
In the world of rapid AI product iterations, best practices and automation are more relevant than ever. Data science is about repeatable experimentation and measured results. Suppose your AI processes can’t be repeated, and production is being done manually. In that case, data science has been reduced to a data hobby.
- Don’t fear failure: deploying AI is a continuous journey
The formula for successful enterprise-wide AI adoption is nurture the idea, plan, prove, improve and then scale. Mistakes will be made, and lessons learned. This is a completely normal – and valuable – part of the process.
Lighthouse projects need to be proven to work, processes need to be streamlined and teams need to upskill. Businesses need a culture of learning and continuous improvement with people at the centre, through shorter cycles, to drive real transformation.
An experimental culture and continuous improvement, through shorter cycles, can drive real transformation. A successful AI strategy acts as a continually evolving roadmap across the different business functions (people, processes and technology) to ensure your chosen solutions are working towards your business objectives. In short, let your business goals guide your AI transformation, not the other way around.
Iron Mountain releases 7-steps to ensure digitisation delivers long-term benefits
Iron Mountain has released practical guidance to help businesses future-proof their digital journeys. The guidance is part of new research that found that 57% of European enterprise plan to revert new digital processes back to manual solutions post-pandemic.
The research revealed that 93% of respondents have accelerated digitisation during COVID-19 and 86% believe this gives them a competitive edge. However, the majority (57%) fear these changes will be short-lived and their companies will revert to original means of access post-pandemic.
“With 80% still reliant on physical data to do their job, now is a critical time to implement more robust, digital methods of accessing physical storage,” said Stuart Bernard, VP of Digital Solutions at Iron Mountain. “Doing so can enhance efficiency and deliver ROI by unlocking new value in stored data through the use of technology to mine, review and extract insight.”
When COVID-19 hit, companies had to think fast and adapt. Digital solutions were often taken as off-the-shelf, quick fixes – rarely the most economical or effective. But they are delivering benefits – those surveyed reported productivity gains (27%), saving time (20%), enhancing data quality (13%) and cutting costs (12%).
So what now?
The Iron Mountain study includes guidance for how to turn quick-fixes into sustained, long-term solutions. The seven-steps are designed to help businesses future-proof their digital journeys and maximize value from physical storage:
1) Gather insights: The COVID-19 pandemic allowed organisations to test and learn. Companies should ensure these insights are fed into developing more robust solutions.
2) Use governance as intelligence: Information governance and compliance are fundamental to data handling. But frameworks aren’t just a set of rules, they hold valuable insights that can be turned into actionable intelligence. Explore your framework to extract learnings.
3) Understand your risk profile: A key early step is to analyse where you are most vulnerable. With data in motion and people working remotely, which records are at risk? What could be moved into the cloud? Are your vendors resilient?
4) Focus where you will achieve greatest impact: To prioritise successfully, you need to know where you will achieve the largest impact. This involves looking beyond initial set-up costs towards the holistic benefits of digitisation, including reducing time spent on manual scanning, and the risk of compliance violations.
5) Reach out and collaborate: We are all in this together. Your IT, security, compliance and facility management teams are all facing the same challenges. Ensure you collaborate across functions to develop robust, integrated solutions.
6) Find a provider who can relate to your digital journey: For companies that still rely heavily on analogue solutions, digitisation can be daunting and risky. It pays to find a vendor who has been on the same journey, understands your paper processes and can guide you through the digital world.
7) Prioritise and evolve communication and training programmes: To reap the full rewards from any digitisation initiative, thorough and continuous communication and training is critical. Encouragingly, our survey found that 81% of data handlers have received training to work digitally which is an excellent step in the right direction, but consider teams beyond data handling to truly succeed.
The research was commissioned by Iron Mountain in collaboration with Censuswide. It surveyed 1,000 data handlers among the EMEA region. It found that the departments that have digitised more due to COVID-19 include IT support (40%), customer relationship management (36%), and team resource planning (34%).
3D Secure: Why are fraudsters still slipping through the net?
By Tim Ayling, VP EMEA, buguroo
There is a constant tension between keeping online payments secure, and offering an easy and frictionless user experience. Digital transformation – especially accelerated by the global pandemic – leaves consumers expecting online services to be seamless. Customers are even liable to abandon a process altogether if they encounter a hurdle.
Financial regulation and security protocols exist to help ensure that a balance is maintained between offering customers this frictionless experience, and keeping them and their funds safe from fraud attacks.
What is 3D Secure?
3D Secure is one such protocol. This payer authentication system is designed to keep card-not-present (CNP) ecommerce payments secure against online fraud. The card issuer uses 3D Secure when a card is used to pay for something online, authenticating the customer’s identity based on personal identifiers, such as the three-digit CVV code on the back of a card, as well as the device they’re using to make the payment and their geolocation or IP address.
3D Secure is important because although transactions can be accepted or denied based on the level of risk, it’s not always as clear as ‘risky’ or ‘not risky’. A small number of transactions will have an undetermined or questionable level of risk attached to them. For example, if a legitimate customer appears to be using a new device to buy goods online, or appears to be attempting to make the transaction from an irregular location. In these instances, 3D Secure provides a step-up authentication, such as asking for a one-time password (OTP).
Getting the right balance
3D Secure is a helpful protocol for card issuers, as it allows banks to comply with Strong Customer Authentication as required by EU financial regulation PSD2 as well as increase security for transactions with a higher level of risk – thereby better filtering the genuine cardholders from fraudsters.
This means that the customers themselves are better protected against fraud, and the extra security helps preserve their trust in the bank to be able to keep their money safe. At the same time, the number of legitimate customers who have their transactions denied is minimised, improving the customer’s online experience.
So why are fraudsters still slipping through the net?
Fraudsters are used to adapting to security protocols designed to stop them, and 3D Secure is no exception. The step-up authentication that is required by 3D Secure in the instance of a questionable transaction often takes the form of an OTP, a password or secret answer known only by the bank and the customer. However, there are various ways that fraudsters have devised to steal this information.
The most common way to steal passwords is through phishing attacks, where fraudsters pretend to be legitimate brands, such as banks themselves, in order to dupe customers into giving away sensitive information. Fraudsters can even replace the pop-up windows that appear to legitimate customers in the case of stepped-up authentication with their own browser windows disguised as the bank’s. Unwitting customers then enter the password or OTP and effectively hand it straight over to the fraudsters.
Even when an OTP is sent directly to a customer’s phone, fraudsters have found a way to intercept this information. They do this through something called a ‘SIM swap scam’, where they impersonate their victim and manage to get the legitimate cardholder’s number switched onto a different SIM card that they own, thereby receiving the genuine OTP in the cardholder’s place.
This is especially an issue for card issuers when taking into account the liability shift that is attached to using 3D Secure. When a transaction is authenticated using 3D Secure, the liability moves to lie with the card issuer, not the vendor or retailer. If money leaves a customer’s account and the transaction was verified by 3D Secure, but the customer says they did not authorise the transaction, the card provider becomes liable for any refunds.
How AI and Behavioral Biometrics can be used to plug the gap
Banks need to find a way to accurately block fraudsters while allowing genuine customers to complete online payments. AI can be used alongside behavioural biometrics as an additional layer of security to cover the gaps in security through continuous authentication of the customer.
Behavioural biometrics can collect and analyse data from thousands of parameters around user behaviour such as their typing speed and dynamics, or the trajectory on which they move the mouse, throughout the entire online session. AI processes are used to dynamically compare this analysis against the user’s usual online profile to identify even the smallest of anomalies, as well as against profiles of known fraudsters and typical fraudster behaviour. AI then delivers a risk score based on this information to banks in real time, enabling them to root out and block the fraudulent transactions.
As this authentication occurs invisibly, the AI technology can recognise if the customer is who they say they are – and that it isn’t a fraudster trying to input a genuine OTP they have managed to steal through phishing or SIM swapping – without adding any additional friction.
Card issuers cannot decline all questionable transactions without losing customers, while approving them without additional checks poses security issues that can result in financial losses as well as losses in customer trust. Behavioural biometrics is a foundational technology that can work simultaneously to 3D Secure to keep customers’ online payments safe from fraud while maintaining a frictionless experience and minimising the risk of chargeback liability for banks.
Motivate Your Management Team
A management team, typically a group of people at the top level of management in an organization, is a team...
The Income Approach Vs Real Estate Valuation
The Income approach is only one of three main classifications of methodologies, commonly referred to as valuation approaches. It’s particularly...
How To Create A Leadership Philosophy
A leadership philosophy describes an individual’s values, beliefs and principles that they use to guide a business or organization. Your...
How to Build an AI Strategy that Works
By Michael Chalmers, MD EMEA at Contino Six steps to boosting digital transformation through AI In the age of artificial...
Leumi UK appoints Guy Brocklehurst to property finance team as Relationship Manager
Multi-specialist bank announces the appointment of Guy Brocklehurst to its property finance team Guy Brocklehurst has joined London-based Leumi UK...
Three times as many SMEs are satisfied than dissatisfied with COVID-19 support from their bank or building society
More SMEs are satisfied (38%) than dissatisfied (13%) with their COVID-19 banking support Decline in SMEs using personal current accounts...
Tax administrations around the world were already going digital. The pandemic has only accelerated the trend.
By Emine Constantin, Global Head of Accoutning and Tax at TMF Group. Why do tax administrations choose to go digital?...
Time for financial institutions to Take Back Control of market data costs
By Yann Bloch, Vice President of Product Management at NeoXam Brexit may well be just around the corner, but it is...
An outlook on equities and bonds
By Rupert Thompson, Chief Investment Officer at Kingswood The equity market rally paused last week with global equities little changed...
Optimising tax reclaim through tech: What wealth managers need to know in trying times
By Christophe Lapaire, Head Advanced Tax Services, Swiss Stock Exchange This has been a year of trials: first, a global...