Testing

Category: ACT

  • Customer Lifetime Value (CLTV) for a Digital Wallet Business

    Customer Lifetime Value (CLTV) for a Digital Wallet Business

    Customer Lifetime Value (CLTV) is a metric that measures the total revenue/profit that a customer is expected to bring over the course of their lifetime with the business. It is an important metric that calculates the worth of all customers. CLTV models can then help address issues such as

    “We have lost this high-value customer, how much can I afford to spend to win them back?” or

    “Which customers should I be targeting to maximize my return on investment?”

    Here are some reasons why understanding and knowing CLTV is essential to a business:

    1. It helps in targeting profitable customers: When the CLTV of each customer is known, it helps in identifying customers who need to be retained or targeted for more products/services and which customers can be let go if there is a risk of churn.

    2. It helps in understanding the kind of customers a business is acquiring: It is always desirable that a business acquires as many as high-value customers, so to evaluate this, calculating the CLTV of new customers relative to existing customers is important. If the CLTV of new customers is higher than that of existing customers, it is a positive for the business as this indicates that the new customers are of high value and profitable customers.

    3. Campaign Effectiveness: CLTV can also be used as a proxy to determine the effectiveness of campaigns where the purpose is to increase the value or revenue from customers. For instance, if a cross-sell or up-sell campaign is being done on a set of customers, and if the CLTV of those customers goes up in the near term, then we can reasonably say that the campaign has been successful.

    4. Business performance: CLTV can also be used as a business performance metric since it essentially tells us the value of all customers, which in turn tells us the value of the business. So, if CLTV increases over time, it is a positive sign for the business.

    5. ROI on Customer acquisition cost (CAC): For a business, (CLTV: CAC) ratio is a critical business metric. This tells us how much value the customer is bringing to the business for every dollar spent on acquiring the customer. This ratio tells you how profitable a customer will be over their lifetime. CLTV: CAC ratio can also yield insights into how efficiently the sales and marketing team are spending money to acquire customers.

    Calculating Customer Lifetime Value

    There are different ways to calculate based on different business models. Here, we will look at calculating CLTV for a digital wallet business.

    CLTV is calculated as follows:

    CLTV = ((T*AOV) AGM)) ALT

    There are four components when it comes to calculating CLTV:

    Component Definition Calculation
    T Average monthly transactions  No Of Transactions/ No of Active Months
    AVPT Average Value Per Transaction Total Transaction Value/ No of Transactions
    ALT Avg Customer Lifespan (In Months) 1/churn probability
    AGM Average gross margin (Revenue – Costs)/Revenue

    Out of these 4 components, Customer lifespan is somewhat difficult to calculate. If customer churn data is available for a longer period, i.e., 8-10 years, then we can arrive at customer lifespan using 1/ (average churn rate).

    Our Methodology

    We at Subex built a CLTV solution for a digital wallet company as part of our campaign intelligence offering.

    Here is the process we followed:

    • We took Active 90 subscribers as our base for this solution and calculated T, AVPT, and AGM at the monthly level using the last 90 days of data.
    • To calculate Average customer lifespan (ALT), we relied upon our churn probability prediction model, which was already integrated with our campaign intelligence solution, and thus, we arrived at customer lifespan. Since we wanted to reduce skewness in churn probability, we decided to create customer segments using RFM and then took the median of churn probability for each segment and used that to calculate the average customer lifespan.
    • We calculated Recency, Frequency, and Monetary scores and created a composite score by assigning weights to each score.

    RFM Composite Score: 60%(Monetary) +20%(Frequency) + 20%(Recency)

    • Using RFM composite scores, we created RFM segments based on percentiles with the following logic. For example, a very low segment contains customers with RFM scores between 0 and 15th percentile.
    RFM Segmentation Segmentation Logic
    Very Low 0-15th Percentile
    Low 15th – 30th Percentile
    Medium 30th – 45th Percentile
    Medium High 45th – 60th Percentile
    High 60th – 75th Percentile
    Very High 75th – 90thPercentile
    Elite Above 90th Percentile
    • For each of these segments, median churn probability scores were calculated, which were then used to calculate the average customer lifespan. This gave us all the components for the calculation of CLTV.

    Integrating Customer Retention Module

    We also built a customer retention module basis CLTV scores of customers. As mentioned earlier, CLTV provides insights on which set of customers’ needs to be prioritized for retention.

    The process of building this module was as follows:

    • We first created risk buckets based on the churn probabilities.
    Risk Bucket Distribution
    Low Risk 0-50%
    Medium Risk 50-70%
    High Risk 70-90%
    Very High Risk 90-100%

    So, a customer having a churn probability of less than 50% will belong to the Low-Risk bucket.

    • Using our RFM segmentation and risk buckets, we created a matrix to prioritize customers for retention.
    RFM Segmentation Vs Risk Bucket Low Risk Medium Risk High Risk Very High Risk
    Very Low P5 P4 P3 P3
    Low P4 P4 P3 P3
    Medium P4 P3 P3 P3
    Medium High P4 P2 P2 P2
    High P3 P2 P1 P1
    Very High P3 P2 P1 P1
    Elite P2 P2 P1 P1

    For prioritization, we created the order as follows: P1>P2>P3>P4>P5

    • Since a customer belonging to either high-risk or very high-risk segment and high, very high, and Elite RFM segment should be prioritized first in a retention campaign, we tagged them as P1. We then continued to move down the prioritization order for customers with lower risk and lower value.
    • When spending on campaign promotions to retain customers, the strategy should be to spend more on highly valuable customers first and then decrease the spending amount as we go down the prioritization order. We created a spending band for each of the priority groups as follows.
    Campaign Spend Bucket Lower Limit Spend Upper Limit Spend
    P1 20% 25%
    P2 15% 20%
    P3 10% 15%
    P4 5% 10%
    P5 0% 5%

    So, suppose a retention campaign is being run on customers in the P1 category. In that case, the maximum amount we can spend on the campaign is 25% of their respective CLTV, so even if we can successfully retain any of the customers, we will still make 3-4 times of our campaign spends.

    Thus, by using CLTV, we created an end-to-end campaign management solution.

    It is also essential to know some of the pitfalls of CLTV because at the end of the day, CLTV is a prediction; therefore, caution is necessary when using it as a guide for making decisions.

    Pitfalls of CLTV:

    1. CLTV cannot be used to justify campaign expenditure, and it is only good as its assumptions.
    2. CLTV is not immune to changes in the macro environment. If inflation is high or there is some geo-political risk, CLTV cannot reflect it immediately.

    Get a 360-degree view of customers in a powerful all-in-one pane for monitoring and managing the complete life cycle of the customers

    Request a demo now!

  • Identifying Suspicious Subscriber Activities for Mobile Wallet Providers

    Identifying Suspicious Subscriber Activities for Mobile Wallet Providers

    Telecoms are privy to fraud from different avenues. One avenue is subscribers acting as agents and performing forex buying and selling activities. Identifying, investigating, and addressing such behaviour is extremely important from a regulatory, anti-money laundering, subscriber protection, and revenue standpoint.

    What’s suspicious and what’s not?

    Telecom operators often have merchants and agents operating across their network for legitimate activities like conducting recharges, selling airtime, and more. To weed out malicious agents from the real ones, CSPs must first replicate the behaviour of agents to define what is ‘suspicious activity.’ This is difficult because simply conducting high-value transactions is not always a red flag. There could be many individual subscribers who move large sums for personal reasons.

    Coming up with a threshold for suspicion is another grey area. This sometimes depends on the economic climate within a specific region and hence requires some business and regional understanding.

    There is typically no way to separate informal merchants using their subscriber lines to receive payments and fraudulent subscribers. Activity patterns across both these groups are similar. Perhaps the only underlying difference is that informal merchants deal with products and services, whereas suspicious subscribers deal in foreign currency in exchange for mobile money.

    Addressing these challenges needs a unique approach because of the nuances in segregating genuine activity from suspicious ones. In the example below, Subex used a mix of industry, market, and business understanding to configure a solution that helped a CSP stay ahead of suspicious subscriber activity and related fraud.

    How Subex did it: A real-world example 

    A major communications service provider with a renowned mobile wallet services platform wanted to stay on top of its subscriber activity to discern suspicious behaviour so that immediate corrective action could be taken. This was important for the CSP to comply with regulatory terms and safeguard the business from malicious forces. It would also support Combating of Financing of Terrorism (CFT) capabilities and foster positive brand perception.

    The operator sensed that certain subscribers were acting as agents and performing unscrupulous money transfer activities. They wanted to identify, investigate, and address such behaviour on priority.

    They chose an approach that delineated behaviour based on specific traits that correlated to suspicious activity. These three behaviour types were:

    • High financial activity like transferring unusually high values
    • Connection density like finding subscribers transacting with an unusually high number of subscribers
    • Volatility or a surge in financial activity or connectedness of a subscriber

    Subex was brought in to implement one of its proprietary solutions to help the CSP get insights into these three categories of subscriber activity.

    Two modules were created – one to identify subscribers with suspicious activity and another to assist the investigations with relevant data points on all suspicious subscribers. Considering there were nearly 4.5 million nodes on the network and approximately 45 million Edges, the task was a difficult one.

    In a nutshell, Subex performed the following actions:

    • Modelled cash selling behaviour using techno-analytical rigour
    • Identified each subscriber’s connections across the massive network using a graph theory-based degree centrality model to discern legitimate transactions from suspicious ones.
    • Investigated transaction behaviour to identify subscribers whose transacted values diverged from usual

    The values and connection density across the three defined behavioural buckets were finalized using a combination of data-driven exploratory analysis and business acumen. Rather than choosing fixed, rigid values as the threshold for suspicion, Subex configured the tool with flexible threshold options that could be custom-set. Thresholds were then assigned to maximize true positives and minimize false positives. These values were used to narrow down the suspicious subscriber base.

    Results from the analysis were shared with the compliance team, giving them a 360-degree view of the subscribers with the main investigation markers. This included subscriber value segments, location details, device information, connection, and value profiling, and national ID reuse details, among others.

    Soon, the solution gained popularity with the telco, and Subex upgraded it with an automated, scalable, flexible, and democratized front-end for ready access for users across the organization. The tool is helping set custom thresholds of suspicion for value transfers and connection density, get the distribution for the reason of suspicion, and capture markers for investigation of these subscribers.

    In a span of two months, the solution identified 161,400 subscribers that were possibly acting as agents. It also gave the CSP valuable insights such as:

    • 64,800 subscribers increased values while 49,200 increased connections. These could possibly be agents who were barred and now operating on subscriber lines.
    • 12,460 subscribers from this base have reused national IDs and can be considered highly suspicious.
    • 40% of suspicious subscribers are concentrated in the capital city, which is also where the concentration of agents is the highest.
    • 47,500 subscribers use basic/feature phones and have suspicious behaviour. These could possibly be vendors and can be considered low-risk.

    Benefits of subscriber activity analytics

    In the short term, performing subscriber activity analysis helps telecom operators periodically weed out subscribers acting as agents and improves how they manage regulatory expectations. In the medium to long term, it discourages fraudsters from building networks to deal in foreign currency. It also saves telecom operators from incurring heavy regulatory fines.

    In summary, fraudulent understanding activity across subscribers often requires a niche approach founded on strong technical expertise and a sound understanding of the telecom industry and local market forces. All of these factors blend to create environments ripe for fraud. Mitigating such risk is up to telecom providers that ought to leverage well-informed strategies as well as technical expertise in terms of data analysis, dashboarding, and automation.

    Mobile wallet provider identified suspicious subscribers with Analytics

    Download the case study to learn more

  • Optimize Insurance Underwriting for customer retention using mobile wallets for an insurance provider

    Optimize Insurance Underwriting for customer retention using mobile wallets for an insurance provider

    When it comes to insurance, the tech-savvy customer of today is no longer encumbered by lengthy, manual, and time-consuming processes where insurance agents must authenticate identity, underwrite risk, set a premium, and then draw up a policy. Traditional insurers are being outpaced by digital-first insurance start-ups that promise mobile-first experience, faster disbursals, self-service capabilities, and process transparency.

    In such a landscape, competition is tough, riding on price. The rise of insurance aggregators allows customers to compare different insurance providers based on price, coverage, claim settlement ratios, benefits, and more. With price being a purchase driver, insurers need to re-examine pricing effectiveness to ensure they are attractive to customers while ensuring that the enterprise remains profitable.

    Why is pricing so important?

    Traditionally Insurance companies have used a cost-oriented pricing model based on claims experience and are calculated from internal data sources. An expected profit margin is added to arrive at the final cost. The issue with this approach is that it doesn’t factor in changing market dynamics like competition, price sensitivity of customers, and economic conditions of the geography that they are operating in.

    Consumers today are also more informed than before. Internet is empowering customers with data to compare different insurance products by price, value, and benefits. Also, with the emergence of direct players and aggregators, prices have been further pushed down.

    With all these changes happening, the approach towards Insurance pricing must be a data-driven approach which means that insurance players should make a significant investment towards their digital infrastructure. This could include:

    Accurate Data Collection: This includes not just internal sources but external sources as well.

    Data Processing Capabilities: Data collected needs to be processed faster, which can help derive consumer insights. These insights then can help in either validating or changing our approach.

    Data Security: Storing the collected data on secure servers and having strict policies on its usage.

    Technological Innovation is rapidly changing the pricing structures across industries, and therefore Insurance companies must adapt and make pricing consumer-centric rather than cost-centric to retain competitive advantage.

    Pricing premiums correctly is critical because it hedges against risk or losses that the policyholder may incur. Moreover, this risk must be diversified across their product portfolio to stay profitable. Underwriting risk involves complex statistical models that ingest different variables, make assumptions about customer behavior, and provide certain outputs that inform the decision on premium costs. However, there are many cases where insurers end up charging customers either too little or too much for the perceived risk. Both of these have a negative impact: Charging the customer too little exposes the insurer to cost liability, impacting profitability. Charging the customer too much leads to customer dissatisfaction and negative brand reputation, impacting long-term profitability. Further, customers are privy to pricing discrepancies, which are easily exposed through insurance aggregators that allow feature-based policy and premium comparisons.

    Given this, insurers must devise more innovative ways to assume and calculate risk to optimize their pricing strategies. A data-driven pricing strategy can significantly enhance pricing efficiencies by enabling more accurate predictability. It will also allow insurers to perform better customer segmentation and risk underwriting, making price a competitive differentiator.

    Case study: Modelling mobile wallet data for customer-centric pricing

    A client in the insurance industry wanted to introduce a new policy plan lower than their existing entry-level plan. It was essential that the premium of the new policy was affordable to cater to customers in the low-income category. The aim was to position this new policy competitively to grow their customer base, and hence the insurer wanted to get the new price point just right.

    Subex designed an innovative approach using mobile wallet data to give the insurer the right insights to make the best decision on pricing. First, data on mobile wallet usage was obtained securely from a third party and used to uncover behavioral patterns. Then, an analysis including medians and percentiles was conducted to arrive at a revenue gain calculator. Assuming various conversion rates, these inputs helped ascertain the ideal pricing for the new policy.

    The primary customer behavior investigated was the maintenance of sufficient monthly balances. The premise was that those able to sustain their balance (after paying monthly bills) would be better positioned to afford a premium for insurance. Based on the outstanding monthly balances, subscribers were categorized into different percentiles groups – 25, 50, 75, 70, 80, and so on. For instance, a balance of $10 within the 25th percentile group meant that 75% of subscribers maintained a balance above $10 for more than 15 days.

    Data from a single month was extracted and grouped into five parameters, i.e., minimum balance, maximum balance, 25th percentile, 50th percentile, and 75th percentile. These five parameters made up the five-point table that was used to design the revenue gain calculator. The revenue gain calculator provided estimations of the monthly revenue gain for the insurer based on different price points. It also provided insights into the expected adoption by the corresponding subscriber groups.

    The analysis yielded an ideal price point of $32. This price point was arrived at using the 70th percentile of wallet balances. Based on the model, it is predicted that 50% of the subscribers would maintain a $32 balance in their wallets for over 30% of the time (9 days).

    The price point analysis conducted by Subex yielded an affordable product for the insurer’s customers and promised some exciting ripple effects. For one, the model has given the organization a reliable manner to underwrite risk by using a new data source, i.e., mobile wallets. It may create some degree of cannibalization as customers from high-value plans downgrade to the lower ones. Finally, it will prevent customer churn and improve retention by allowing customers to switch to a new plan instead of changing their insurance provider. Ultimately, the price point analysis is helping the insurer effectively price newer policies for greater profitability.

    Mobile Wallet Provider Identified Suspicious Subscribers with Analytics

    Learn More

  • From Default to Debt: Stemming the Flow with Bad Debt Analytics

    From Default to Debt: Stemming the Flow with Bad Debt Analytics

    Telecommunications customer service teams have long been tasked with multiple roles – delivering superior experiences, being savvy with multiple tools, working with bots or assisted automatons, and curating omni-channel interactions to name a few. They are also responsible for ensuring that customers complete payments and clear their dues on time.

    While many solutions are available to enhance the operational efficiency of contact centers, this particular role – managing bad debt – requires a different approach, and here’s why. Globally, the incidents of fraud and defaulters are increasing, which put customer service teams on the backfoot as they struggle to make finance-related decisions. like:

    When late payments morph into bad debts

    The words ‘bad debt’ began as a way for banks and credit card companies to refer to customers who defaulted on their loan or credit card payments. Today, the telecommunications industry borrows this term from finance for postpaid and enterprise customers who have not cleared their dues as per their billing cycle. Bad debt is a real challenge within the telecommunications industry and a contributor to annual revenue losses. Bad debts can be as high as 2% of the total revenue for telecom operators.

    Billing to collection lifecycles in telecom are a bit different from the banking and lending industry. Banking and credit firms often deal with chronic defaulters through one-time settlements or loan restructuring. Telecommunication players, on the other hand, have a different play. Since initial debt recovery falls within the realm of customer service teams, the agents execute gradual actions like sending recurring payment reminders and barring outgoing calls. They often have no way of knowing whether a late payment is a genuine one or whether it is nefarious, leaning towards non-payment. Finally, as the default risk becomes severe, the job is outsourced to an external vendor – a third party collection agency – whose job it is to recover the payment or ‘bad debt’.

    The question is: Is there a way to predict the risk of bad debt way ahead of time?

    Here too, telecom can take a leaf out of the lessons and examples demonstrated by finance. Many global banks and lending agencies are investing in ways to capture 360-degree customer data in a bid to create large datasets that feed intelligent risk scoring models. Such models use atypical customer data to forecast the likelihood of credit risk and credit fraud. These predictions are then leveraged to make smarter decisions about lending, like levying higher interest rates, thereby bringing down the rates of default.

    Bad debt analytics

    An analysis of the data collected during each stage of billing to collections lifecycle will yield key insights. The five main stages are:

    Stage 1: Bill generation

    Stage 2: Pay-by-date

    Stage 3: Early delinquency stage (time of default > 30 days)

    Stage 4: Advanced delinquency or bad debt (time of default is between 60-120 days)

    Stage 5: Collection, disconnection or legal action (time of default > 120 days)

    Analytics models can ingest this lifecycle data to reveal patterns and correlations that provide a probability score of defaulting to bad debt. This entire process is known as bad debt analytics. It focuses on, among other things, minimizing how many customers move from Stage 2 to Stage 3. As seen in the image, stage 3 is the critical stage that separates late payments from defaulters.

    Tools to facilitate early detection of bad debt

    Bad debt analytics are a constellation of modern tools and technologies that can greatly assist customer service teams in making the right decision at the right time when late payment instances arise. It can also provide a host of correlations that inform onboarding about the propensity to default. Some of these tools:

    • Automated data capture – Telecom databases are replete with voluminous customer information that contain key insights. But first, this data lying in disparate systems across sales, marketing, finance, and network teams must be integrated into a single repository. Through automated data capture using connectors and APIs, Telcos can unlock data siloes, mechanizing the job of real-time data integration. This integrated data is what feeds bad debt analytics models for relevant output.
    • Machine learning models – These models accelerate the data-to-insight lifecycle by crunching data, understanding trends, and finding patterns that cannot be perceived by humans. An example is how one of our models uncovered that a majority of the defaulters in the handset sales department of a leading Middle East operator were migrant customers. This insight helped the telco adapt its credit processes to promote appropriate repayment behaviors and lower its bad debt risk.
    • Smart dashboards – Output from the models must be displayed in the right manner to the customer service teams so they can take or recommend the right actions at the right time. For instance, high risk of default in a specific location may indicate poor KYC practices by salespeople. Knowing this, the operator can immediately halt new sales transactions in that area and redirect staff for training.

    Learn how our Analytics Centre of Trust can help your business

    Schedule Demo

  • Midiendo la verdadera rentabilidad: un enfoque holístico y granular

    Midiendo la verdadera rentabilidad: un enfoque holístico y granular

    En el mercado actual de rápida evolución pero sobresaturado, la industria de las telecomunicaciones ha cambiado su forma de trabajar. Ha comenzado a fusionarse o cruzarse con otras industrias, como los medios y los servicios financieros, ofrecer más variantes de productos y servicios personalizados a clientes valiosos, y adoptar rápidamente nuevos avances técnicos y adaptarse a las últimas ofertas, como 5G. Como resultado, el proceso comienza a involucrar a varios departamentos en la estrategia central y la toma de decisiones. Pero cada departamento monitorea su propio conjunto de KPI, que están aislados de los demás. Como resultado, los KPI más fáciles de monitorear como los ingresos, la base de suscriptores, el número de instalaciones y las actualizaciones del sitio tienen prioridad, mientras que el concepto de ganancias y márgenes se ve afectado y, por lo tanto, la responsabilidad de múltiples elementos de costo. Para superar este problema se requiere un ajuste en la estrategia general que implica la colaboración entre departamentos y un cambio en los criterios de éxito del crecimiento de la línea superior (basado en los ingresos) al crecimiento de la línea de fondo (basado en las ganancias) con la presencia de todos los interesados ​​en el área de ganancias y ganancias. Vista de pérdida (P&L) de la empresa, lo que resulta en una única versión de la verdad.

    Enfoque

    Un enfoque holístico y granular es un intento de medir la verdadera rentabilidad de los clientes, productos, activos de red y regiones geográficas mediante la adopción de un enfoque más estructurado y metódico. Todo el proceso se puede dividir en cuatro pasos, como se muestra a continuación.

    1. Comprender: las empresas tienen la mayor parte de la información necesaria para calcular las ganancias, pero se encuentra dispersa en varios sistemas y departamentos. En este paso, se realizan reuniones con diferentes partes interesadas para analizar los modelos comerciales existentes y comprender sus gastos generales de ingresos y costos, la metodología actual de cálculo de márgenes y las fuentes de datos disponibles.

    2. Desarrollar: una vez que se identifican todas las partidas relacionadas con los ingresos y los costes, se desarrolla una lógica

    a. para alinear los elementos en categorías significativas. Por ejemplo: ingresos por uso (dentro / fuera del producto), ingresos por interconexión (nacionales / internacionales), costo de la red (por ejemplo, alquileres / energía), seguros (teléfonos / servicios administrados), etc.

    b. definir el modelo de costos para determinar el costo unitario. Por ejemplo, el costo total del sitio de la red se divide por el uso total del sitio de la red (duración) para derivar un costo por segundo, y luego el costo se multiplica para obtener el costo de la red por transacción. De la misma manera, los ingresos de un suscriptor se asignan de forma inversa a los sitios de la red en la proporción de uso del suscriptor.

    3. Procesar y analizar: una vez que se desarrolla la lógica, se procesan y analizan diferentes elementos de línea de ingresos y costos al nivel granular posible. Se instala el motor de hardware y computación de última generación, que combina múltiples fuentes de diferentes sistemas y perfiladores en ejecución para obtener el margen final de una manera programada.

    4. Calcular: Este paso implica la configuración de un sistema con características de roll-up y slice-dice para derivar el margen final contra diferentes parámetros. Las reglas se definen para incluir o excluir las líneas de pedido según el tipo de nivel. Por ejemplo, el margen se puede calcular con diferentes parámetros como Cliente, Producto, Segmento, Estado, etc., como se muestra a continuación.

    Fig. Ilustración de la vista de pérdidas y ganancias del cálculo del margen.

    Casos de uso

    Algunos de los casos de uso en los que Subex ha entregado utilizando la metodología dada son los siguientes:

    1. Visibilidad de los componentes del costo de la red en la rentabilidad del producto: En el compromiso dado, el equipo de la red quería responsabilidad del equipo del producto al diseñar nuevos productos mientras el equipo de la red se esfuerza por cumplir con los requisitos de la red para los productos. Subex creó una contabilidad detallada de vista de pérdidas y ganancias en todos los componentes de costo de la red para los productos en función de su uso y puntos de precio de corte derivados para los productos que se ofrecen en los servicios ofrecidos. Pudimos identificar algunos productos con altos ingresos por ventas, pero con un margen negativo o poco profundo después de las asignaciones de costos de la red. Todo el proceso trajo más transparencia, buscando la coordinación entre ambos departamentos.

    2. Identificación de clientes altamente rentables: En este compromiso, el cliente quería cambiar los criterios de los clientes más valiosos, de ingresos a ganancias. Subex ayudó al cliente a comprender la contribución exacta que cada cliente hizo al resultado final. El cliente reconoció que el método de ingresos había sido engañoso e inexacto, ya que dos clientes con ingresos similares pueden tener costos variables. Además, ayudó al cliente a identificar los componentes clave del costo, convirtiendo a los clientes de altos ingresos en clientes de bajo margen.

    Conclusión

    En una situación en la que los precios son cada vez más difíciles de intervenir, optimizar las ofertas de costos y servicios se vuelve esencial para impulsar los márgenes y requiere redefinir el criterio del éxito desde los ingresos y las ventas hasta la rentabilidad. La vista de pérdidas y ganancias holística y granular propuesta aumenta la transparencia, lo que genera más información procesable y una mejor comprensión de la estructura de costos, lo que ayuda a identificar los impulsores de la causa raíz.

    Vea cómo Subex Analytics Center of Trust puede ayudar a su empresa.

    Programe una demostración

  • What are the 5 ways that Business Intelligence Helps Telecommunications Providers?

    What are the 5 ways that Business Intelligence Helps Telecommunications Providers?

    Business Intelligence helps any organization transform the raw data into meaningful insights which can be leveraged by the organization to make business decisions such as CAPEX/OPEX investments, which products to launch, insights on customer behavior and much more. The importance of Business Intelligence in telecommunication has grown more than ever with the continuous growth in data as telcos ventured into new domains or launch new services. Here are the 4 ways that business intelligence can telecommunication drive more value-

    Customer & Campaign Intelligence: As telecom operators serves millions of customers, gaining individual customer level insights is very critical to provide personalized services that customers demand these days. Business Intelligence can provide that intelligence to operators by converting the raw data into consumable information. As a result, telcos can do targeted marketing campaigns, refine their pricing strategy and develop or launch products based on empirical consumer understanding.

    Proactive Customer Service: A good customer service can have a significant impact on reducing your customer churn. Customer issues and complaints are always going to be a challenge for telecom operators.  A Business Intelligence solution can provide process improvements to provide support and service. Customer care representatives can up their service for the most common customer complaints, finding stats on the complaints that customers waited on the phone longest about or even identifying support questions by email that took them the longest to respond.

    Network Intelligence– Telco customers are consuming more data than ever. They are streaming videos, playing online games, communication with colleagues or family/friends through video calls. This can cause network congestions, service interruptions resulting in loss of customer and revenue. Also, as 5G and IoT become mainstream, the high bandwidth and low latency requirements will require telcos to have clear visibility of their network performance tracking the activities happening at cell site level.

    Revenue Intelligence

    A Business Intelligence solution can help you gather information around revenue composition across different dimensions such as cell-sites, products/services, devices, geographies, customer segments and more. As a service provider, it is important for you to know the most profitable avenues that is driving the business growth.

    Product Performance

    A Business Intelligence solution can help capture information related to product usage, performance, revenue and margins. Based on the intelligence, operators can bundle their products and services for maximum ROI. The organization can decide to discontinue unprofitable products and launch profitable products.

    Build and Monitor Trusted Business KPIs in Real Time

    Schedule Demo

  • Measuring the True Profitability: A holistic and granular approach

    Measuring the True Profitability: A holistic and granular approach

    In today’s rapidly evolving but oversaturated market, the telecommunication industry has changed its way of working. It has started to merge or intersect with other industries, such as media and financial services, offer more variants of products and personalized services to valuable customers, and quickly embrace new technical advents and adapt to the latest offerings, like 5G. As a result, the process starts involving multiple departments in core strategy and decision making. But each department monitors one’s own set of KPIs, which are in isolation from others. As a result, easier to monitor KPIs like revenue, subscriber base, number of installations, and site upgrades get priority while the concept of profits and margins take a hit and, thus, the accountability of multiple cost items. To overcome this issue requires a rejig in overall strategy involving collaboration between departments and a change in success criteria from top-line growth (revenue-based) to bottom-line growth (profit-based) with a presence of every stakeholder in the Profit & Loss(P&L) view of the company, resulting in a single version of the truth.

    Approach

    A holistic and granular approach is an attempt to measure the true profitability of customers, products, network assets, and geographical regions by adopting a more structured and methodical approach. The whole process can be divided into four steps, as shown below.

    1. Understand: Companies have most of the information needed for profit calculation, but it is scattered across multiple systems and departments. In this step, meetings with different stakeholders are done to analyze existing business models and understand their revenue & cost overheads, current margin computation methodology, and available data sources.

    2. Develop: Once all the line items related to revenue and cost items are identified, a logic is developed

    • a. to align the items into meaningful categories. For example: Usage revenue (within/outside product), Interconnect Revenue (Domestic/International), Network cost (e.g. Rentals/Energy), Insurance (Handset/Managed Services), etc.
    • b. to define cost model for ascertaining the per-unit cost. E.g., total network site cost is divided by overall network site usage(duration) to derive a cost per sec, and then cost is multiplied to obtain the network cost per transaction. Same way, a subscriber’s revenue is reverse mapped to network sites in the ratio of the subscriber’s usage.

    3. Process and Analyse: Once the logic is developed, different revenue and cost line items are processed and analyzed at the granular level possible. The state-of-the-art hardware & computing engine is put in place, combining multiple sources from different systems and running profilers to derive the final margin in a scheduled way.
    4. Compute: This step involves setting up a system with features of roll-up and slice-dice to derive the final margin against different parameters. Rules are defined to include or exclude the line items based on the type of level. E.g., margin can be computed against different parameters like Customer, Product, Segment, State, etc. as shown below.

    Fig. Margin Computation P&L View Illustration.

    Use Cases

    Some of the use cases where Subex has delivered using the given methodology are as follows:

    1. Visibility of network cost components in product profitability: In the given engagement, the network team wanted accountability from the product team while designing new products as the network team tries hard to meet the network requirements for the products. Subex created a detailed P&L view accounting in all network cost components for the products based on their usage and derived cut-off price points for the products given the services offered. We were able to identify some products with high sales revenue, but with negative or shallow margin after network cost allocations. The whole process brought more transparency, seeking coordination between both departments.
    2. Identifying high profitable customers: In this engagement, the client wanted to change the criteria of most valued customers, from revenue to profits. Subex helped the client to understand the exact contribution each customer made to the bottom line. The client acknowledged that the revenue method had been deceptive and inaccurate as two customers having similar revenue can have varying costs. Also, it helped the client identify the key cost components, making high revenue customers into low margin customers.

    Conclusion

    In a situation when prices are becoming harder to intervene, optimizing cost and service offerings become essential to push margins and require re-defining the yardstick of success from revenue & sales to profitability. The proposed holistic and granular P&L view increases transparency leading to more actionable insights and a better understanding of the cost structure, helping identify root cause drivers.

    See how Subex Analytics Centre of Trust can help your business.

    Schedule a demo

  • Cómo Next Best Action puede ofrecer un mayor valor para el cliente – Subex Limited

    Cómo Next Best Action puede ofrecer un mayor valor para el cliente – Subex Limited

    El motor de recomendación o Next Best Action juega un papel esencial en la influencia de las preferencias humanas. Ya sea Amazon, Netflix, YouTube o cualquier otra empresa importante, utilizamos sus recomendaciones en nuestra vida cotidiana. La siguiente mejor acción o el motor de recomendación tiene un impacto significativo en el rendimiento del negocio tanto en términos de generación de ingresos como de satisfacción del cliente.

    Entonces, ¿por qué los operadores necesitan un motor de recomendación? La mayoría de los operadores de telecomunicaciones tienen una base de clientes en millones y cientos de productos. Con tantos productos para combinar con los del cliente, ¿cómo identifica el mejor producto para cada cliente? Y no olvidar las múltiples variables que definen el comportamiento y las necesidades del cliente.

    Las empresas pueden usar el resultado del motor de recomendaciones de múltiples maneras para impulsar su estrategia. Si el plan es aumentar las ventas o mejorar los ingresos, el usuario comercial puede poner una condición adicional en el modelo de que el precio debe ser más alto que el precio actual suscrito. Si el objetivo es la retención, un usuario comercial puede establecer condiciones para que el producto recomendado tenga el mismo precio o menos que el producto que el cliente actual está utilizando. En el caso de la venta cruzada, podemos seguir la misma estrategia que la empresa quiere que los clientes se expongan a nuevos productos.

    Aparte de los beneficios tangibles que un motor de recomendación puede ofrecer a las empresas; Puede proporcionar muchos beneficios intangibles a los clientes, como una mejor experiencia, simplicidad para elegir y satisfacción, lo que lleva a una mayor lealtad del cliente.

    Diferentes enfoques para la próxima mejor acción

    Las técnicas más populares utilizadas son el filtrado colaborativo y el filtrado basado en contenido. El filtrado colaborativo significa filtrar los productos que un usuario podría esperar en función de las acciones de usuarios similares y el filtrado basado en contenido, también conocido como filtrado cognitivo, donde las recomendaciones de productos se realizan en base a una comparación entre el contenido de los elementos y un perfil de usuario. El enfoque de filtrado colaborativo supone que los consumidores con un perfil similar pueden tener las mismas necesidades y tomar decisiones similares, mientras que, en función del contenido, los datos históricos del cliente forman la base de la recomendación.

    Existen muchos otros métodos no tradicionales y mucho más avanzados, como el aprendizaje profundo, el aprendizaje social y la factorización de tensor que se basan en el aprendizaje automático y las redes neuronales para impulsar el proceso de recomendación y llevarlo al siguiente nivel. Creará un proceso mucho más eficiente para CSAT (puntuación de satisfacción del cliente) y retención.

    Construir el motor de recomendaciones no es suficiente

    Los operadores necesitan un objetivo claro al implementar la siguiente mejor estrategia de acción. El motor de recomendación podría recomendar un producto de menor precio como la mejor opción para los clientes que conduce a la pérdida de ingresos para el operador. Por lo tanto, es necesario identificar que el producto que mejor se ajusta cumple con los objetivos del cliente y del operador.

    Además, los productos de telecomunicaciones tienen una validez diferente, lo que significa que un cliente que haya comprado un producto con una validez de 60 días comprará solo después de 60 días. Por lo tanto, el modelo debe tener en cuenta esta brecha de compra al recomendar nuevos productos.

    Un desafío que enfrentamos con uno de nuestros clientes fue que los clientes estaban comprando múltiples productos similares en un mes determinado, lo que dificultaba las recomendaciones. Para superar esto, decidimos que bajo cada tipo de producto: datos, voz, SMS y VAS, recomendaríamos los mejores productos 3-4. Logramos esto al proporcionar clasificaciones para los productos en la canasta de recomendaciones. La clasificación también ofrece flexibilidad al usuario comercial para que pueda agregar más condiciones a los productos en función de su estrategia, como ventas adicionales, ventas agresivas o ventas cruzadas.

    Actualmente estamos logrando un 70% -80% de precisión en los 5 mejores productos recomendados para diferentes segmentos de valor para el cliente.

    ¿Usted necesita La Siguiente Mejor Estrategia de Acción?

    Usar las interacciones o ideas de los clientes es la mejor manera de identificar los patrones de comportamiento de sus clientes y usarlos para proporcionar las mejores recomendaciones a sus clientes. Un motor de recomendaciones puede ayudarlo a identificar los comportamientos de los clientes, los patrones de compra y traducir los conocimientos en un enfoque hiperpersonalizado para sus campañas de marketing. El resultado es una segmentación de usuarios más profunda con claridad, mayor LTV y mayor retención de clientes.

    Aquí hay un punto de vista sobre la importancia de la próxima mejor oferta o motor de recomendación para su negocio.

    Descarga el PoV

  • How Next Best Action can deliver higher customer value

    How Next Best Action can deliver higher customer value

    The recommendation engine or Next Best Action plays an essential part in influencing human preferences. Whether Amazon, Netflix, YouTube, or any other major company, we use their recommendations in our day to day lives. The next best action or the recommendation engine has a significant impact on the business performance both in terms of revenue generation and customer satisfaction.

    So why do operators need Recommendation Engine? Most of the telecom operators have a customer base in millions and hundreds of products. With so many products to match with customer’s, how do you identify the best product fit for each customer? And not to forget the multiple variables that are defining the customer behavior and needs.

    Businesses can use the output from the recommendation engine in multiple ways to drive their strategy. If the plan is to upsell or revenue enhancement, the business user can put an extra condition in the model that the price should be higher than the current subscribed price. If retention is the objective, a business user can set conditions for the recommended product to have the same price or less than the existing product customer is using. In the cross-sell case, we can follow the same strategy as the business wants customer exposure to new products.

    Other than tangible benefits that a recommendation engine can offer to business; it can provide a lot of intangible benefits to customers, like better experience, simplicity to choose, and satisfaction, leading to higher customer loyalty.

    Different approaches for Next Best Action

    The most popular techniques used are collaborative filtering and content-based filtering. Collaborative filtering means filtering out products that a user might expect based on actions by similar users and content-based filtering, also known as cognitive filtering, where product recommendations are made based on a comparison between the content of the items and a user profile. The collaborative filtering approach assumes that consumers with a similar profile may have the same needs and make similar choices while in content-based, the historical customer data form the basis for the recommendation.

    There are many other non-traditional and much more advanced methods such as deep learning, social learning, and tensor factorization that are based on machine learning and neural networks to power the recommendation process and take it to the next level. It will create a much efficient process for CSAT and retention.

    Building the recommendation engine is not enough

    Operators need a clear objective when implementing the next best action strategy. The recommendation engine might recommend a lower price product as the best fit for customers leading to revenue loss for the operator. So, it’s necessary to identify that the best-fit product meets both customer and operator objectives.

    Also, telecom products have different validity, which means a customer who had bought a product with 60 days validity will buy only after 60 days. So, the model must account for this purchase gap while recommending new products.

    One challenge that we faced with one of our clients was that customers were buying multiple similar products in a given month, making recommendations difficult. To overcome this, we decided that under each product type – Data, Voice, SMS & VAS, we would recommend the top 3-4 products. We accomplished this by providing rankings for products in recommendation basket. Ranking also offers flexibility to the business user that they can further add more conditions on products based on their strategy, such as up-sell, aggressive -sell, or cross-sell.

    We are currently achieving 70%-80% accuracy in the top 5 recommended products for different customer value segments.

    Do you need the Next Best Action strategy?

    Using customer interactions or insights is the best way to identify your customer behavior patterns and use them to provide the best recommendations to your customers. A recommendation engine can help you identify customer behaviors, buying patterns, and translate the insights into a hyper-personalized approach for your marketing campaigns. The result is a more in-depth user segmentation with clarity, higher LTV, and higher customer retention.

    Here is a Point of View on the importance of the Next Best Offer or Recommendation Engine for your business.

    Download the PoV

  • Why Analytics Alone is Insufficient for Telcos Seeking Increase in Customer Experience and Profitability?

    Why Analytics Alone is Insufficient for Telcos Seeking Increase in Customer Experience and Profitability?

    The past decade has seen the telecom industry embark upon a momentous growth path. The number of unique mobile subscriptions worldwide has almost doubled from 2.6 billion in 2009[1] to 5.1 billion[2] in 2019, thanks to the quick evolution from 2G to 3G, 4G, and now 5G, along with innovations in IoT, M2M, Artificial Intelligence (AI) and cloud computing. While this growth phase has led to the industry accumulating humongous amounts of data, a majority of them have not reaped the benefits of monetizing the data at their disposal—similar to how Facebook, Google or the many others are successfully doing.

    Data monetization is vital for Telcos to stay ahead

    Telcos are advantaged by a unique position in the ICT value chain. For example, a simple event such as planning a holiday today involves using your smartphone for a variety of tasks—from booking flights and hotels to researching the best places to explore. All of these result in the generation of varied datasets, to which telcos have access to. However, a complicated ecosystem of challenges is preventing telcos from successful internal and external data monetization.

    As Telcos continue to face increasing pressure in their top-line business, internal data monetization can bring them a strategic advantage in solving business problems. From the understanding of revenue trends such as the accurate pinpointing of developments in the voice or data business to gaining deep insight into customer likes and dislikes, data analytics can address several business issues. Similarly, external monetization opens opportunities for telcos in a range of industry verticals.

    However, to stay relevant in the market, telcos will have to stay ahead on the trends, which can lead them to opportunities. Messaging (SMS), which was an exclusive offering of telcos has now been taken up by players like WhatsApp and Facebook Messenger. Employing data analytics to analyze and predict trends is the only way to stay ahead.

    Why analytics alone won’t help?

    Several leading telcos have already realized the power that lies in data analytics as a key strategic pillar and continue to invest in advanced data technologies. However, the ROI for data analytics investments by telcos continues to stand unproven, especially on an incremental basis. The reason: Most telcos view data analytics as a technology asset, leaving open a wide gap between technology and business goals. This severe lack of coordination has resulted in analytics and data science being viewed as a technology solution rather than a business problem, causing business goals to suffer.

    The answer lies in domain-driven analytics

    For telcos seeking to stay competitive, domain-driven analytics is the answer. Spearheaded by deep domain knowledge and wide industry exposure a domain-driven analytics solution understands the requirements of telcos operating in specific regions or market conditions. Through the analysis of existing data sets, domain-driven analytics can transform current stats into future possibilities.

    Take the successful use case of how a telco in a developing market used domain-driven analytics to increase revenue. The telco was using its existing business model of acquiring more customers to increase revenue. Using domain-driven analysis, however, it was found that the current ‘active base’ which was 8.5 days needed to be improved in order to increase revenue. The telco used the recommendations and conducted an integrated campaign which led to the increase of the ‘active base’ from 8.5 to 9.5 days and a boost in revenue by 25%.

    With 25+ years of experience in driving data-driven business transformation for global telcos, Subex has garnered a lot of experience in applying domain-driven analytics for telecom.

    [1] https://www.budde.com.au/Research/Global-Telecoms-The-Big-Picture-2019-Key-Industry-Statistics

    [2] https://www.gsmaintelligence.com/

     

    To find out more about how domain-driven analytics can impact your ROI, Increase Customer Experience and Profitability.

    Download the webinar recording now!