Written by @InterpretableAI & IraIraCohen In the first four parts of our series on correlation analysis, we discussed the importance of this capability in root cause analysis in a number of business use cases, and then specifically in the context of promotional marketing, telco and algorithmic trading. In this blog we walk through how to leverage correlation analysis to address the challenges in ensuring a seamless online payment experience by the end-user. In general, digital payments include:  Online - browser-based purchases of goods/services  In-store - tapping mobile device at a point-of-sale/scanning a code to pay In-app - purchase of goods/services through an app such as features in a mobile game Peer-to-peer - send/receive money through a digital service/platform as exemplified by services such as Venmo, Zelle and the like End-use of digital payment include, but not limited to, BFSI (Banking, financial services and insurance), Healthcare, IT & Telecom, Media & Entertainment, Retail & E-commerce, Transportation.  Unlike matters pertaining to maintaining corporate liquidity, cash is increasingly becoming the not preferred option in the context of consumer-to-business transactions. The COVID-19 crisis and ongoing fears of infection have prompted consumers and businesses to rely, more than ever, on digital and contactless payment options. The infographic below (from McKinsey’s report) highlights the same. In the realm of contactless payment, biometric authentication is expected to gain momentum in the coming years. It is a verification method that involves the biological characteristics of the person. The verifications include facial recognition, fingerprinting scanning, heartbeat analysis, and vein mapping.  Since the founding of PayPal, numerous players have surfaced in the payment space, for example, but not limited to, Square, Stripe, Affirm, Venmo, Chargebee, Zelle, WhatsApp Pay, Novi. As per this whitepaper, 1.7 billion adults globally remain outside of the financial system with no access to a traditional bank, even though one billion have a mobile phone and nearly half a billion have internet access. To this end, payment systems such as Diem, based on blockchain, have been proposed. The Diem Blockchain is a decentralized, programmable database designed to support a low-volatility cryptocurrency that will have the ability to serve as an efficient medium of exchange for billions of people around the world. Novi is a new digital wallet for the Diem payment system. As per the report by Research And Markets, the Global Digital Payment Market size is expected to reach $175.8 billion by 2026, rising at a market growth of 20% CAGR during the forecast period. A sister report by Grand View Research USD breaks down the growth of the global digital payment market size to $236.10 billion by 2028 (a CAGR of 19.4% from 2021 to 2028) by solution type as shown below. Source: Grandview Research Factors driving the growth include, but not limited to, use of smartphones becoming ubiquitous, increasing demand for contactless payments and increasing customer expectations. In addition, factors such as the growing percentage of the global population using banking facilities and the growing adoption of open banking APIs are expected to create new growth avenues for the digital payment market. Headwinds include deteriorating perception of digital payments security over the past year and a growing concern with payments made via social apps and “Internet of Things” devices.” Smooth payment performance is the bedrock of end-user’s e-commerce experience and, consequently, key to contain churn. Performance is no longer limited to an outage. In a report published in October 2020, Gartner highlighted that the traditional KPIs centered around payment performance have been thrown into a tailspin by COVID-19. Going forward, the report calls out the following metrics: Latency: This pertains to the response time in executing a payment. Increase in latency - it can be traced back to a wide variety of reasons such as, but not limited to, gateway or processor issues - could potentially result in timeouts, lost sales and violation of contractual SLAs. Outages (e.g., click here) impacts both the end customer as well as the merchant. Authorization and decline rates: This pertains to how many submitted transactions are approved for payment by the issuer. Typically, approval rates for transactions carried out at the physical point of sale (POS) typically achieve much higher approval rates than the digital commerce counterpart. In light of this, it is important to monitor the metric for POS and digital separately. Also, a sudden shift to digital may warrant a recalibration on the issuer side as well. Anomalies in the digital case can potentially stem from, for example, the fraud detection system being a bit aggressive to contain downside. In an article published last year, Fry highlighted the following flavors of payment acceptance problems: Lack of funds from the payment method                                            “Card not present” transactions                                                  Wrong, missing or expired information                                            Fraud triggers                                                                   Payment requests don’t share the same format                                     Genuine cross-border “foreign” transactions get easily declined                  Currency conversion Fraud detection rates: This pertain to how many submitted transactions successfully make it through the fraud screening tools and are processed for payment. As more and more first-time customers use digital channels, it may result in higher fraud detection rates online. In light of this, it is important to measure the online and offline fraud rates separately - this would help triage the impact on sales and take action accordingly. Payment processing costs: As in-store sales move online, credit and debit card processing fees will apply to a greater portion of your overall sales. Plus, rates are materially less expensive at the physical POS than they are online, thus the credit card processing fees per transaction - which includes interchange fees that are paid to the issuing bank, gateway fees, processor fees, fraud detection costs and more - are likely to increase.  In a similar vein, Kar called out Responsiveness - How quick is the service provider to address issues when a service request is raised by a user? - as one of the key factors impacting user experience in the context of mobile payments - significance value of this hypothesis was found to be less than 0.05 under the model described in therein. Further, the following is recommended: “… enhance the speed at which complaints raised across channels surrounding failures of transactions are addressed ... hugely impact the customer relationship management and impact positively the responsiveness perceived by the customer when any challenge is faced from the use of a digital payment platform.”   Delivery high payment performance is non-trivial. It stems from the complexity of the payments ecosystem as exemplified by the infographics below. Source: GSMA A deep pipeline between the customer and the merchant (refer to the figure below) and potentially a large set of vendors makes the payment experience susceptible to cascading effects. This is akin to the tail latency problem discussed by Dean and Barroso (also see this, this, this). They highlight: “Even rare performance hiccups affect a significant fraction of all requests in large-scale distributed systems.” Source: Deloitte Likewise, one or more factors such as, but not limited to, multiple API handshakes, network hiccups (lost connectivity and/or congestion), availability of one or more intermediaries, directly impact the end-user’s payment experience. Going forward, with blockchain based decentralized architecture gaining momentum, providing resilience against faults would pose an interesting problem. For now, we shall leave the discussion on that front for another time.  Correlation between two or more payment processing KPIs (as discussed in BlueSnap’s article) can help triage the impact to a merchant’s bottomline: Payment Conversion Rate Conversion Rate By Bank Conversion Rate By Payment Method And Card Type   Checkout Abandonment Rate     Landing Page Optimization Rate   Mobile Vs. Desktop Clean Rate (defined as the total percentage of transactions that had a successful outcome) Fraud Rate   Order Rejections Rate  Chargeback Rate (defined as the number of transactions that are disputed by shoppers when they see an unfamiliar charge on their credit card bill) Uptime (what percentage of the day, week, or year is your provider online and processing successfully?) Consider the transactions at an ATM as an example. Potential issues can pertain to, for example, authentication, anomalies in aggregated withdrawal - amount and/or # transactions - per day, abnormally high decline rate owing to the fraud detection system gone wrong. Also, in order to reduce mean time to remediation (MTTR), one may have to slice-and-dice the analysis by, for instance, type of ATM and geography. The two examples below illustrate the importance of anomaly detection and correlations for root cause analysis. In the first example, Anodot’s platform alerts on abnormal drops in payment transaction success rate broken down by payment providers, geography and more. A dip in the success rate for one payment provider in a certain county was correlated to increased errors in one of the payment APIs for one of the banks in that country - this played a key role in rapid remediation of the issue. Interestingly, one would expect the aforementioned to be close to 1.0; however, in practice, we observe that the transaction success rate rises during the day and falls at night (that is, more transactions are successfully completed during the day compared to night time) and are often far from 1.0. The second example comes from monitoring a trading platform. Metrics such as, # transactions and deposits are monitored to surface any potential issues. In the example below, the number of deposits for trading accounts dropped abnormally and was correlated with an increase in latency to the main database handling those transactions - leading to a quick RCA indicating an issue with the DB. Recently, in order to boost the end-user experience, big bank adopted Anodot’s platform for payment and trading monitoring.  To wrap up, complexity of the payments ecosystem  lies at the root of the issues that adversely impact end-user experience. The metrics are diverse and hence, discovering correlations between the metrics can help root cause the issue at hand and drive a reduction in mean time to remediation (MTTR). The benefits to fintech companies include reduction in operational expenses and risk, protection of revenue, and improved customer experience. Continue this series to discover applications of and challenges with applying correlation analysis in: a wider business context ecommerce, specifically promotions fintech, specifically algorithmic trading network performance in telecommunications

For modern businesses faced with increasing volumes and complexity of data, it’s no longer efficient or feasible to rely on analyzing data in BI dashboards. Traditional dashboards are great at providing business leaders with insights into what's happened in the past, but what if they need actionable information in real time? What if they want to use their data to estimate what may happen in the future? Companies are taking notice. In a survey by Deloitte, 67% of executives said they were not satisfied with the benefits of their existing tools and resources. The highest-performing companies use data science empowered by AI/ML to bridge that gap. The Limitations of Dashboards For today's data-driven companies, there are thousands of metrics that can be gathered and collected to gain insights into business performance. As relevant data points grow exponentially, it becomes challenging to use traditional dashboards to track these metrics and make informed timely decisions. Here are some of the most significant limitations of conventional BI dashboards: Lack of real-time anomaly detection prevents proactive incident management Most BI dashboards do not show data in real-time, and when they do, there are so many metrics cluttering screens that users can easily miss the most critical information. Timely intervention is crucial to modern businesses, which often run tightly integrated ecosystems of applications and infrastructure that stretch across multiple departments and process enormous amounts of data. For example, leading adtech platform, Rubicon Project, fields trillions of bid requests per month and needs to analyze data points from millions of potential sources. In an environment like that, every minute can have significant impact. They found that traditional dashboards failed to deliver the real-time detection and response capability necessary to intervene before anomalies impacted their bottom line. According to Gartner, downtime costs the average business more than $300K per hour. Enterprises need the capability to manage these systems proactively rather than reactively. Over-reliance on historical data Most companies configure and use traditional dashboards to track KPIs and other critical business metrics to understand how their business and systems perform. One factor often missed by decision-makers is that the data they view in traditional dashboards describes what has already happened and might not be a reliable indicator of what will happen in the future. Moving from descriptive to predictive modes of thinking requires a deep understanding of the business context and critical thinking, which can be challenging for any person, or even a dedicated team, given the diversity of the data set, new trends, and fluctuating behaviors. Missing small incidents that have a negative impact Some incidents are hard to spot, but that doesn't mean they won't significantly impact the business. When undetected, hard-to-spot incidents can accumulate and can end up having the same impact as more prominent issues. A typical scenario involves incidents affecting only one business component. These isolated issues can easily get lost in KPIs based on a calculated average of multiple metrics. For example, a server cluster might be displaying a 99.99% average uptime. If one server in that cluster is experiencing an anomalously high amount of downtime, it could remain invisible to the dashboard. A single server is a small data point in a data center with thousands of servers, but it could be vital depending on what that server is running. CEO dashboards lack correlation When it comes to business intelligence and gaining real, actionable insights, choosing which metrics to include is more art than science. CEO dashboards only answer the questions that users who configure the system think to ask, but actionable insights can be present in any metric. This limitation is even more problematic because some insights can only be correlated when data is considered across multiple metrics, even if there is no apparent connection. For data to be actionable, it needs to have a holistic view of all relevant information and the impacts of the decision across the business. CEO dashboards fall short when missing data links delay or lead to misinformed decisions that harm the organization. Cluttered dashboards and false positives Sometimes, even with all the necessary information, BI dashboards struggle to present a coherent picture. With CEO dashboards, in particular, there's some guesswork in determining ahead of time what information is important enough to display in the limited real estate available on the screen. When alerts start to pop up, it can be difficult to tell which data is necessary or worth ignoring. The sheer volume and increasing complexity of data can quickly overwhelm the dashboard interface, making it much harder for business leaders to consume in a timely, accurate manner. Lack of intelligent prioritization Collecting thousands of events or alerts every minute from your applications and infrastructure, and presenting that data in a dashboard isn’t analytics. Users apply filters on this data, performing their own analysis and work. Deriving intelligence from data shouldn’t require an end user to define what to look for, or where, or what are the most critical KPIs, or what normal or abnormal is. This is not intelligence because a user is telling the dashboard exactly what data to show. Leverage the Power of AI Analytics Business strategy is only effective if empowered with enough intelligence and agility to outmaneuver the competition. Traditional dashboards don't provide insight fast enough in today's data-driven world, and when a business can lose hundreds of thousands of dollars in a single hour due to a pricing glitch on an e-Commerce site, the stakes are too high. Companies need real-time, actionable insights across all data metrics relevant to performance. The best-performing businesses leverage BI solutions empowered by AI and machine learning to eliminate the need for human correlation across the millions of critical metrics needed to understand business and system performance. Grouping and correlating multiple anomalies by design, Anodot's AI-powered analytics elevates the essential insights first. By learning the normal behavior of millions of metric’s, Anodot detects only the most impactful incidents and alerts relevant teams at the start. In the example above, an online business was alerted immediately to a spike in cost for a Google Ad campaign. It was an incident so granular that it would have been overlooked for days using traditional monitoring tools. As an AI analytics solution with anomaly detection capabilities, Anodot can monitor millions of metrics at a granular level, giving both the detail and scale needed to identify the business incidents that matter. Automated anomaly detection and contextualized alerts break free talented specialists and business leaders from the pain of manually monitoring dashboards so they can step in when it counts.

When organizations track metrics by the thousands, millions, or even billions, it’s helpful in many ways to understand which metrics have close relationships, meaning when one metric behaves in a certain way, one or more additional metrics can be expected to behave in a similar or opposite way. What is Correlation Analysis? Correlation analysis calculates the level of change in one variable due to the change in the other. If there is shown to be a strong correlation between two variables or metrics, and one of them is being observed acting in a particular way, then you can conclude that the other one is also being affected in a similar manner. Finding relationships between disparate events and patterns can reveal a common thread, an underlying cause of occurrences that, on a surface level, may appear unrelated and unexplainable. A high correlation points to a strong relationship between the two metrics, while a low correlation means that the metrics are weakly related. A positive correlation result means both metrics increase in relation to each other, while a negative correlation means that as one metric increases, the other decreases. Why Correlation Analysis is Important Correlation analysis can reveal meaningful relationships between different metrics or groups of metrics. Information about those connections can provide new insights and reveal interdependencies, even if the metrics come from different parts of the business. If there is shown to be a strong correlation between two variables or metrics, and one of them is being observed acting in a particular way, then you can conclude that the other one is also being affected in a similar manner. This helps to group related metrics together to reduce the need for individual processing of data. The Benefits of Correlation Analysis Reduce Time to Detection In anomaly detection, working with a vast number of metrics and surfacing correlated anomalous metrics helps draw relationships that not only reduce time to detection (TTD) but also supports shortened time to remediation (TTR). As data-driven decision-making has become the norm, early and robust detection of anomalies is critical in every industry domain, as delayed detection adversely impacts customer experience and revenue. Reduce Alert Fatigue Another important benefit of correlation analysis in anomaly detection is in reducing alert fatigue by filtering irrelevant anomalies (based on the correlation) and grouping correlated anomalies into a single alert. Alert storms and false positives are significant challenges faced by organizations today – getting hundreds, even thousands of separate alerts from multiple systems, when many of them stem from the same incident. Reduce Costs Correlation analysis helps significantly reduce the costs associated with the time spent investigating meaningless or duplicative alerts. In addition, the time saved can be spent on more strategic initiatives that add value to the organization. Example Use Cases for Correlation Analysis Marketing professionals use correlation analysis to evaluate the efficiency of a campaign by monitoring and testing customers’ reactions to different marketing tactics. In this way, they can better understand and serve their customers. Financial planners assess the correlation of an individual stock to an index such as the S&P 500 to determine if adding the stock to an investment portfolio might increase systematic risk of the portfolio. For data scientists and those tasked with monitoring data, correlation analysis is incredibly valuable when used for root cause analysis and reducing time to detection (TTD) and time to remediation (TTR). Two unusual events or anomalies happening at the same time/rate can help to pinpoint an underlying cause of a problem. The organization will incur a lower cost of experiencing a problem if it can be understood and fixed sooner rather than later. Technical support teams can reduce the number of alerts they must respond to by filtering irrelevant anomalies and grouping correlated anomalies into a single alert. Tools such as Security Information and Event Management (SIEM) systems do this automatically to facilitate incident response. How Anodot Uses Correlation of Metrics in Business Monitoring Business monitoring is the process of collecting, analyzing, and using metrics and key performance indicators (KPIs) to track an organization’s progress toward reaching its business objectives and to guide management decisions. Anomaly detection is a key method for identifying when a business process is experiencing an unexpected change that may indicate an underlying issue is derailing the process. As organizations become more data-driven, they find themselves unable to scale their analytics capabilities without the help of automation. When an organization has thousands of metrics (or more), analyzing individual metrics can obscure key insights. A faster method is to use machine learning based correlation analysis in order to group related metrics together. In this way, when a metric becomes anomalous, all the related events and metrics that are also anomalous are grouped together in a single incident. This helps to reduce data processing time, reveal the root cause of an incident, and tie events together to reduce alert fatigue. On average, customers using Anodot have found correlation analysis helps reduce alert noise by up to 99%. An Example of Correlation in Business Monitoring Consider the applicability of correlation analysis in the realm of eCommerce promotions.  For many retailers, the last quarter of the year accounts for more than 50 percent of their annual sales. Most merchants run various promotions to boost sales that correspond with Black Friday, Cyber Monday, and other holiday-related events. Multiple factors are at play with any promotion, including the promotion type, promotional pricing, audience targeting, purchase intent, timeliness, media used for the promotion, and numerous other factors. Correlation analysis finds a natural fit to determine which factor(s) play a key role in driving the top and bottom lines in the sales. The ability to identify strong correlations would help marketers double down on the corresponding promotions. To illustrate, consider the figure below, which shows how two correlated anomalies – a spike in page views (top chart) and add to carts (bottom chart) – for an eCommerce site points to an anomalous sales pattern. The shaded area (the baseline) is the normal pattern of sales for a promotional event of this nature. Clearly, the add to cart metric is underperforming. Correlating the relevant event (the sale) and the related metrics (pageviews and add to cart) together, it underscores the irregularity of a drop in both those metrics. When the event started, the team was alerted about the fact that the sales event did not yield the expected increase in both the correlated metrics; in fact, page views actually dropped(!) 46 percent compared to the expected spike, leading to a drop of 66 percent in add to cart. These drops were identified because the effect of the sales event (an “external” variable to the metric), was correlated to the values of the metric. If the correlation between the metrics and the event was not taken into account, the drop would have seemed like an increase. Discovering the relationships among data metrics has many practical applications in business monitoring. Correlation analysis can help identify the root cause of a problem and vastly reduce the time to remediate the issue. It also helps to group events together in order to reduce the number of alerts generated by the events, in turn reducing alert fatigue among support personnel and the cost of investigating duplicative alerts.

Anodot’s payment gateway analytics provides clear visibility into the payments environment to enable the fast detection of transaction performance issues, anomalies and trends so that no revenues or customers are lost

Autonomous monitoring creates the real-time visibility that is critical for enabling faster detection of transaction issues and optimized approval rates and revenues

Leading CSP use AI-based autonomous monitoring of granular CX, performance and telemetry data to provide a flawless user experience by proactively mitigating incidents and service degradation

CSPs opting to adopt an AI-based approach to network monitoring need to balance time to value and return on investment, while securing the best possible solution for their specific needs. Here’s what you need to know.

When forecasting cloud costs, accuracy is key. So how did Amazon Forecast fare against Anodot? Read on for the results.

In 2020, the gaming market generated over 177 billion dollars, marking an astounding 23% growth from 2019. While it may be incredible how much revenue the industry develops, what’s more impressive is the massive amount of data generated by today’s games.  The Enormous Data Challenge in Gaming There are more than 2 billion gamers globally, generating over 50 terabytes of data each day. The largest game companies in the world can host 2.5 billion unique gaming sessions in a single month and host 50 billion minutes of gameplay in the same period.  The gaming industry and big data are intrinsically linked. Companies that develop capabilities in using that data to understand their customers will have a sizable advantage in the future. But doing this comes with its own unique challenges.  Games have many permutations, with different game types, devices, user segments, and monetization models. Traditional analytics approaches, which rely on manual processes and interventions by operators viewing dashboards, are insufficient in the face of the sheer volume of complex data generated by games.  Unchecked issues lead to costly incidents or missed opportunities that can significantly impact the user experience or the company's bottom line. That's why many leading gaming companies are turning to AI and Machine Learning to address these challenges. Gaming Analytics AI  Gaming companies have all the data they need to understand who their users are, how they engage with the product, and whether they are likely to churn. The challenge is gaining valuable business insights into the data and taking action before opportunities pass and users leave the game.  AI/ML helps bridge this gap by providing real-time, actionable insights on near limitless data streams so companies can design around these analytics and act more quickly to resolve issues. There are two fundamental categories that companies should hone in on to make the best use of their gaming data: The revenue generating opportunities in the gaming industry is one reason it’s a highly competitive market. Keeping gamers engaged requires emphasizing the user experience and continuous delivery of high-quality content personalized to a company's most valued customers.  Customer Engagement and User Experience  Graphics and creative storylines are still vital, and performance issues, in particular, can be a killer for user enjoyment and drive churn. But with a market this competitive, it might not be enough to focus strictly on these issues.  Games can get an edge on the competition by investing in gaming AI analytics to understand user behaviors, likes, dislikes, seasonality impacts and even hone in on what makes them churn or come back to the game after a break.  AI-powered business monitoring solutions deliver value to the customer experience and create actionable insights to drive future business decisions and game designs to acquire new customers and prevent churn. AI-Enhanced Monetization and Targeted Advertising All games need a way to monetize. It's especially true in today's market, where users expect games to always be on and regularly deliver new content and features. A complex combination of factors influences how monetization practices and models enhance or detract from a user's experience with a game.  When monetization frustrates users, it's typically because of aggressive, irrelevant advertising campaigns or models that aren't well suited to the game itself or its core players. Observe the most successful products in the market, and one thing you will consistently see is highly targeted interactions.  Developers can use metrics gleaned from AI analytics combined with performance marketing to appeal to their existing users and acquire new customers. With AI/ML, games can use personalized ads that cater to users' or user segments' behavior in real-time, optimizing the gaming experience and improving monetization outcomes.  Using AI based solutions, gaming studios can also quickly identify growth opportunities and trends with real-time insight into high performing monetization models and promotions. Mobile Gaming Company Reduces Revenue Losses from Technical Incident  One mobile gaming company suffered a massive loss when a bug in a software update disrupted a marketing promotion in progress. The promotion involved automatically pushing special offers and opportunities for in-app purchases across various gaming and marketing channels. When a bug in an update disrupted the promotions process, the analytics team couldn't take immediate action because they were unaware of the issue.  Their monitoring process was ad hoc, relying on the manual review of multiple dashboards, and unfortunately, by the time they discovered the problem, it was too late. The result was a massive loss for the company – a loss of users, a loss of installations, and in the end, more than 15% revenue loss from in-app purchases. The company needed a more efficient and timely way to track its cross-promotional metrics, installations, and revenue. A machine learning-based approach, like Anodot's AI-powered gaming analytics, provides notifications in real-time to quickly find and react to any breakdowns in the system and would have prevented the worst of the impacts. Anodot's AI-Powered Analytics for Gaming  The difference between success and failure is how companies respond to the ocean of data generated by their games and their users. Anodot's AI-powered Gaming Analytics solutions can learn expected behavior in the complex gaming universe across all permutations of gaming, including devices, levels, user segments, pricing, and ads. Anodot's Gaming AI platform is specifically designed to monitor millions of gaming metrics and help ensure a seamless gaming experience. Anodot monitors every critical metric and establishes a baseline of standard behavior patterns to quickly alert teams to anomalies that might represent issues or opportunities. Analytics teams see how new features impact user behavior, with clear, contextual alerts for spikes, drops, purchases, and app store reviews without the need to comb over dashboards trying to find helpful information.  The online gaming space represents one of the more recent areas where rapid data collection and analysis can provide a competitive differentiation. Studios using AI powered analytics will keep themselves and their players ahead of the game.