Subscription-based services are becoming more popular in today’s society. Therefore, any company that engages in the subscription-based business needs to understand the user behavior and minimize the number of users canceling their subscription, i.e. minimize churn. According to marketing metrics, the probability of selling to an existing user is markedly higher than selling to a brand new user. Nonetheless, it is of great importance that more focus is directed towards preventing users from leaving the service, in other words preventing user churn. To be able to prevent user churn the company needs to identify the users in the risk zone of churning. Therefore, this thesis project will treat this as a classification problem. The objective of the thesis project was to develop a statistical model to predict churn for a subscription-based service. Various statistical methods were used in order to identify patterns in user behavior using activity and engagement data including variables describing recency, frequency, and volume. The best performing statistical model for predicting churn was achieved by the Random Forest algorithm. The selected model is able to separate the two classes of churning users and the non-churning users with 73% probability and has a fairly low missclassification rate of 35%. The results show that it is possible to predict user churn using statistical models. Although, there are indications that it is difficult for the model to generalize a specific behavioral pattern for user churn. This is understandable since human behavior is hard to predict. The results show that variables describing how frequent the user is interacting with the service are explaining the most whether a user is likely to churn or not. / Prenumerationstjänster blir alltmer populära i dagens samhälle. Därför är det viktigt för ett företag med en prenumerationsbaserad verksamhet att ha en god förståelse för sina användares beteendemönster på tjänsten, samt att de minskar antalet användare som avslutar sin prenumeration. Enligt marknads-föringsstatistik är sannolikheten att sälja till en redan existerande användare betydligt högre än att sälja till en helt ny. Av den anledningen, är det viktigt att ett stort fokus riktas mot att förebygga att användare lämnar tjänsten. För att förebygga att användare lämnar tjänsten måste företaget identifiera vilka användare som är i riskzonen att lämna. Därför har detta examensarbete behandlats som ett klassifikations problem. Syftet med arbetet var att utveckla en statistisk modell för att förutspå vilka användare som sannolikt kommer att lämna prenumerationstjänsten inom nästa månad. Olika statistiska metoder har prövats för att identifiera användares beteendemönster i aktivitet- och engagemangsdata, data som inkluderar variabler som beskriver senaste interaktion, frekvens och volym. Bäst prestanda för att förutspå om en användare kommer att lämna tjänsten gavs av Random Forest algoritmen. Den valda modellen kan separera de två klasserna av användare som lämnar tjänsten och de användare som stannar med 73% sannolikhet och har en relativt låg missfrekvens på 35%. Resultatet av arbetet visar att det går att förutspå vilka användare som befinner sig i riskzonen för att lämna tjänsten med hjälp av statistiska modeller, även om det är svårt för modellen att generalisera ett specifikt beteendemönster för de olika grupperna. Detta är dock förståeligt då det är mänskligt beteende som modellen försöker att förutspå. Resultatet av arbetet pekar mot att variabler som beskriver frekvensen av användandet av tjänsten beskriver mer om en användare är påväg att lämna tjänsten än variabler som beskriver användarens aktivitet i volym.

This past decade, the majority of services have been digitalized and data more and more available, easy to store and to process in order to understand customers behaviors. In order to be leaders in their proper industries, subscription-based businesses must focus on their Customer Relationship Management and in particular churn management, that is understanding customers cancelling their subscription. In this thesis, churn analysis is performed on real life data from a Software as a Service (SaaS) company selling an advanced cloud-based business phone system, Aircall. This use case has the particularity that the available dataset gathers customers data on a monthly basis and has a very imbalanced distribution of the target: a large majority of customers do not churn. Therefore, several methods are tried in order to diminish the impact of the imbalance while remaining as close as possible to the real world and the temporal framework. These methods include oversampling and undersampling (SMOTE and Tomek's link) and time series cross-validation. Then logistic regression and random forest models are used with an aim to both predict and explain churn.The non-linear method performed better than logistic regression, suggesting the limitation of linear models for our use case. Moreover, mixing oversampling with undersampling gives better performances in terms of precision/recall trade-off. Time series cross-validation also happens to be an efficient method to improve performance of the model. Overall, the resulting model is more useful to explain churn than to predict it. It highlighted some features majorly influencing churn, mostly related to product usage. / Under det senaste decenniet har många tjänster digitaliserats och data blivit mer och mer tillgängliga, enkla att lagra och bearbeta med syftet att förstå kundbeteende. För att kunna vara ledande inom sina branscher måste prenumerationsbaserade företag fokusera på kundrelationshantering och i synnerhet churn management, det vill säga förståelse för hur kunder avbryter sin prenumeration. I denna uppsats utförs kärnanalys på verkliga data från ett SaaS-företag (software as a service) som säljer ett avancerat molnbaserat företagstelefonsystem, Aircall. Denna fallstudie är speciell på så sätt att den tillgängliga datamängden består av månatlig kunddata med en mycket ojämn fördelning: en stor majoritet av kunderna avbryter inte sina prenumerationer. Därför undersöks flera metoder för att minska effekten av denna obalans, samtidigt som de förblir så nära den verkliga världen och den tidsmässiga ramen. Dessa metoder inkluderar översampling och undersampling (SMOTE och Tomeks länk) och korsvalidering av tidsserier. Sedan används logistisk regression och random forests i syfte att både förutsäga och förklara prenumerationsbortfall. Den icke-linjära metoden presterade bättre än logistisk regression, vilket tyder på en begränsning hos linjära modeller i vårt användningsfall. Dessutom ger blandning av översampling med undersampling bättre prestanda när det gäller precision och återkoppling. Korsvalidering av tidsserier är också en effektiv metod för att förbättra modellens prestanda. Sammantaget är den resulterande modellen mer användbar för att förklara bortfall än att förutsäga dessa. Med hjälp av modellen kunde vissa faktorer, främst relaterade till produktanvändning, som påverkar bortfallet identifieras.

Churn prediction is used to identify customers that are becoming less loyal and is an important tool for companies that want to stay competitive in a rapidly growing market. In retail, a dynamic definition of churn is needed to identify churners correctly. Customer Lifetime Value (CLV) is the monetary value of a customer relationship. No change in CLV for a given customer indicates a decrease in loyalty. This thesis proposes a novel approach to churn prediction. The proposed model uses a Recurrent Neural Network to identify churners based on Customer Lifetime Value time series regression. The results show that the model performs better than random. This thesis also investigated the use of the K-means algorithm as a replacement to a rule-extraction algorithm. The K-means algorithm contributed to a more comprehensive analytical context regarding the churn prediction of the proposed model. / Illojalitet prediktering används för att identifiera kunder som är påväg att bli mindre lojala och är ett hjälpsamt verktyg för att ett företag ska kunna driva en konkurrenskraftig verksamhet. I detaljhandel behöves en dynamisk definition av illojalitet för att korrekt kunna identifera illojala kunder. Kundens livstidsvärde är ett mått på monetärt värde av en kundrelation. En avstannad förändring av detta värde indikerar en minskning av kundens lojalitet. Denna rapport föreslår en ny metod för att utföra illojalitet prediktering. Den föreslagna metoden består av ett återkommande neuralt nätverk som används för att identifiera illojalitet hos kunder genom att prediktera kunders livstidsvärde. Resultaten visar att den föreslagna modellen presterar bättre jämfört med slumpmässig metod. Rapporten undersöker också användningen av en k-medelvärdesalgoritm som ett substitut för en regelextraktionsalgoritm. K-medelsalgoritm bidrog till en mer omfattande analys av illojalitet predikteringen.

Användarchurn är ett stort problem för mobile applikationer, och speciellt för älsoapplikationer. Eftersom mobila applikationer är en så ny industri finns det nästan ingen forskning om churn i mobila applikationer, och ingen forskning alls om fokuserad på churn i hälsoapplikationer. Syftet med den här studien var att undersöka vilka dimensioner ett företag med en mobil hälsoapplikation måste ta hänsyn till för att analysera churn, samt att ge dem verktyg för att själva analysera churn i framtiden. Studien gjordes i for m av en case - studie på företaget Lifesum. En omfattande   litteraturstudie genomfördes för att skapa ett initia l ramverk för att analysera churn i en hälsoapplikation. Intervjuer gjordes med både användare samt anställda på Lifesum, med frågor baserade på det initiala ramverket. Resultatet av intervjuerna användes sen för att förbättra ramverket, för att ge en mer korrekt bild av churn i en mobil hälsoapplikation. Resultatet användes också för att skapa en sammanställning av churn på case - företaget Lifesum. Studien resulterade I ett ramverk med flera olika dimensioner, med flera olika kopplingar. De imensioner som främst påverkar churn är användarnöjdhet samt bytesbarriärer, och de är i  in tur influerade av andra faktorer. Analyses av churn i Lifesum - applikationen visar att churn är komplext. Användare har flera olika syften med att använda applikationen, samt ännu fler anledningar till varför de slutar använda appen. Ett diagram som kartlägg er de olika användarna presenteras, men en analys om hur applikationen behöver förändras för att passa behoven av de olika användargrupperna. En  rekommendation ges också till företag att undersöka vilken typ an användare de vill fokusera på, baserat både på vad de kan göra samt vilken riktning de vill att applikationen ska ta. / User churn is a big problem for mobile applications, and in particular for healthapplications. Because mobile applications is such a new industry there is almost no research on churn in mobile applications, and none at all regarding health. The purpose of this research was to explain what dimensions a mobile health application company need to take into account when analyzing churn, and provide them with the tools to analyze churn in the future.The study was done in the form of a case study at the mobile health company Lifesum. An extensive literature study was conducted to create an initial framework for analyzing user churn in a health application. Interviews were conducted with both users and with Lifesum employees, with questions based on the initial framework. The results of the interviews were then used to augment the initial framework, to represent a more accurate image of user churn in a mobile health application. The results of the interviews were also used to create an overview of churn in the specific case study of the Lifesum app.The outcome of the study was a framework with many different dimensions, with intricate connections between them. The main dimensions influencing user churn are user satisfaction and switching barriers, and they are in turn influenced by other factors. The analysis of churn in the Lifesum application shows that churn is complex. Users have many different purposes for using the app, and even more reasons for why they stop using the application. A diagram mapping the different users is presented, with analysis regarding how the app needs to change to cater to the different user groups. A recommendation is also given for companies to investigate which type of users they want to cater to, based on both what they can do and what direction they want the application to go.

Population churn--the population turnover experienced by a community--can have differential effects on a community. Mass society theory suggests that because the churn rate experienced by communities can contribute to their uprooting, fragmentation, and isolation, churn is a potent threat to the stability of our modern day communities. Social capital theory, to the contrary, suggests otherwise. Social capital theory suggests that churn can have positive effects on communities by bringing new migrants with valuable human capital skills and experiences to communities. These migrants bring to their new communities the potential for creating new jobs, spurring economic development, and for initiating housing starts that expand housing options for the poor and minorities. In so doing, they help create and sustain vibrant, growing modern day communities. Yet in spite of the significant role churn may play in determining the health and viability of modern day communities, it has been overlooked in the migration literature, which is mostly dominated by individual-level research on the causes and effects of migration, particularly the pecuniary benefits to movers. Using county-level data and multivariate analyses, this research seeks to fill this gap in the literature by examining the relationship between the community and churn, from the perspectives provided by social capital and mass society theories.

Customer churn has always been a problem to be addressed by the telecommunication service providers. So far, work done in this regard was based on analyzing historical data of the customers by using different data mining techniques. Investigations based on individual user behavior with a motive of churn prediction are expected to give an idea about the user’s point view towards churn. Data volumes/data usage of the users is seen as parameter to assess the satisfaction of the users with the service. The subjective and objective behavior of the mobile phone users has been captured by collecting data about the data volumes/data usage for both Wi-Fi and mobile services along with their ratings of Quality of Experience (QoE).   The Experience Sampling Method has been deployed to collect the user data. Android tool was used to collect weekly data volumes of the users. A questionnaire was prepared with questions regarding quality, annoyance and churn risk of the users. The questionnaire was used to collect the weekly opinions of the users on the service. A total of 22 users participated in the study, of which 3 persons churned to other service provider during the study. The data collected in the study was analyzed using averages, correlations and decision trees. Comparisons were made between Wi-Fi and mobile services, churners and non-churners/active users. A 2-fold churn prediction model was proposed based on conclusions of the study.

Customer churn is always a grievous issue for the Telecom industry as customers do not hesitate to leave if they don’t find what they are looking for. They certainly want competitive pricing, value for money and above all, high quality service. Customer churning is directly related to customer satisfaction. It’s a known fact that the cost of customer acquisition is far greater than cost of customer retention, that makes retention a crucial business prototype. There is no standard model which addresses the churning issues of global telecom service providers accurately. BigData analytics with Machine Learning were found to be an efficient way for identifying churn. This thesis aims to predict customer churn using Big Data analytics, namely a J48 decision tree on a Java based benchmark tool, WEKA. Three different datasets from various sources were considered; first includes Telecom operator’s six month aggregate active and churned users’ data usage volumes, second includes globally surveyed data and third dataset comprises of individual weekly data usage analysis of 22 android customers along with their average quality, annoyance and churn scores by accompanying theses. Statistical analyses and J48 Decision trees were drawn for three different datasets. From the statistics of normalized volumes, autocorrelations were small owing to reliable confidence intervals, but confidence intervals were overlapping and close by, therefore no much significance could be noticed, henceforth no strong trends could be observed. From decision tree analytics, decision trees with 52%, 70% and 95% accuracies were achieved for three different data sources respectively.      Data preprocessing, data normalization and feature selection have shown to be prominently influential. Monthly data volumes have not shown much decision power. Average Quality, Churn Risk and to some extent, Annoyance scores may point out a probable churner. Weekly data volumes with customer’s recent history and necessary attributes like age, gender, tenure, bill, contract, data plan, etc., are pivotal for churn prediction.

O mercado de jogos eletrônicos online tem crescido em ritmo acelerado nos últimos anos, particularmente a partir do surgimento do modelo de negócio baseado em serviços. Como consequência, as publicadoras destes jogos passaram a compartilhar problemas comuns na área de serviços, como a erosão do lucro causada pela evasão de usuários. Modelos preditivos têm sido utilizados no combate à evasão em mercados como os de telefonia móvel e de cartões de crédito, setores que detêm um grande volume de informações demográficas e econômicas a respeito dos seus consumidores. Já os publicadores de jogos muitas vezes só possuem o endereço eletrônico dos jogadores. O objetivo deste trabalho é propor um modelo de previsão de evasão com base exclusivamente nos padrões de acesso de usuários em jogos online, onde estes registros temporais são submetidos a um conjunto de operadores que analisam os dados no domínio do plano tempo-frequência, utilizando a Transformada Discreta de Wavelet. Sua principal contribuição está na proposta de parametrização dos dados de entrada para classificadores probabilísticos baseados no algoritmo k-Nearest Neighbors. Testados com dados reais de acessos de usuários ao longo de alguns meses em um jogo online, os classificadores foram avaliados com o uso de curvas ROC (Receiver Operating Characteristic) e de elevação. A abordagem proposta nesta tese, baseada na análise no domínio do plano tempo-frequência, apresentou resultados satisfatórios. Não apenas superiores se comparados com as abordagens no domínio do tempo ou da frequência, mas também comparáveis aos desempenhos encontrados por modelos com centenas de variáveis preditivas utilizados em outros mercados. / The online gaming market has rapidly grown in recent years, particularly since the rise of the service-based business model. As a result, the publishers of these games have started to share usual problems from the services business, like the profit erosion caused by customer churn. Predictive models have been used to address the churn problem in the mobile phones and credit cards markets, where companies have a huge volume of demographic and economic data about their customers. While game publishers often have only their users email addresses. The goal of this study is to propose a model for churn prediction based solely on the online games users access patterns, where these time entries are fed into a set of operators that are able to analyze the data in the time-frequency plane domain, using the Discrete Wavelet Transform. Its main contribution is the input data parameterization proposed for the probabilistic classifiers based on the k-Nearest Neighbors algorithm. Tested with real data from an online game users access over a few months, the classifiers were evaluated using ROC (Receiver Operating Characteristic) and lift curves. The approach proposed in this thesis, based on the analysis of the time-frequency plane domain, has shown satisfactory results. Not only higher when compared with approaches based on both time or frequency domains, but also comparable to performances found on models with hundreds of predictive variables used in other markets.

This thesis considers the problem of churn within the electricity distribution sector. More specifically, this study evaluates how supervised machine learning can be used by a Swedish electricity distributor in order to identify customer churn. The data was by provided by the electricity distributor and covered personal, geographical and contract specific information regarding the company’s customers. The provided data was complemented with external data covering the customers’ financial positions. Based on this information the possibility to predict customer churn over a three-month period with a gradient boosted decision tree was evaluated. The results from the proposed models suggests that the possibility to identify customer churn is rather poor and could not be used in a practice. This is believed to be a result of unbalanced class distributions and that the data provided simply is not informative enough to accurately predict customer churn. If more information about the customers is collected, with predictive analyses in mind, the performance of the model is likely to increase.

O mercado de jogos eletrônicos online tem crescido em ritmo acelerado nos últimos anos, particularmente a partir do surgimento do modelo de negócio baseado em serviços. Como consequência, as publicadoras destes jogos passaram a compartilhar problemas comuns na área de serviços, como a erosão do lucro causada pela evasão de usuários. Modelos preditivos têm sido utilizados no combate à evasão em mercados como os de telefonia móvel e de cartões de crédito, setores que detêm um grande volume de informações demográficas e econômicas a respeito dos seus consumidores. Já os publicadores de jogos muitas vezes só possuem o endereço eletrônico dos jogadores. O objetivo deste trabalho é propor um modelo de previsão de evasão com base exclusivamente nos padrões de acesso de usuários em jogos online, onde estes registros temporais são submetidos a um conjunto de operadores que analisam os dados no domínio do plano tempo-frequência, utilizando a Transformada Discreta de Wavelet. Sua principal contribuição está na proposta de parametrização dos dados de entrada para classificadores probabilísticos baseados no algoritmo k-Nearest Neighbors. Testados com dados reais de acessos de usuários ao longo de alguns meses em um jogo online, os classificadores foram avaliados com o uso de curvas ROC (Receiver Operating Characteristic) e de elevação. A abordagem proposta nesta tese, baseada na análise no domínio do plano tempo-frequência, apresentou resultados satisfatórios. Não apenas superiores se comparados com as abordagens no domínio do tempo ou da frequência, mas também comparáveis aos desempenhos encontrados por modelos com centenas de variáveis preditivas utilizados em outros mercados. / The online gaming market has rapidly grown in recent years, particularly since the rise of the service-based business model. As a result, the publishers of these games have started to share usual problems from the services business, like the profit erosion caused by customer churn. Predictive models have been used to address the churn problem in the mobile phones and credit cards markets, where companies have a huge volume of demographic and economic data about their customers. While game publishers often have only their users email addresses. The goal of this study is to propose a model for churn prediction based solely on the online games users access patterns, where these time entries are fed into a set of operators that are able to analyze the data in the time-frequency plane domain, using the Discrete Wavelet Transform. Its main contribution is the input data parameterization proposed for the probabilistic classifiers based on the k-Nearest Neighbors algorithm. Tested with real data from an online game users access over a few months, the classifiers were evaluated using ROC (Receiver Operating Characteristic) and lift curves. The approach proposed in this thesis, based on the analysis of the time-frequency plane domain, has shown satisfactory results. Not only higher when compared with approaches based on both time or frequency domains, but also comparable to performances found on models with hundreds of predictive variables used in other markets.