--- title: Cognitive Model | customer journey map description: customer journey map | Machine learning - big data - artificial intelligence - data mining - analytics --- Posts about # customer journey map ## [Insight from chatbots and WhatsApp with NLP](https://blog.sinnetic.com/en-us/cognitive_services/insight-from-chatbots-and-whatsapp-with-nlp) Posted by [Gabriel Contreras](https://blog.sinnetic.com/en-us/cognitive_services/author/gabriel-contreras) | Jun 23, 2021 3:56:34 PM As organizations open new client interaction channels, unstructured information capture grows... [ CONTINUE READING ](https://blog.sinnetic.com/en-us/cognitive_services/insight-from-chatbots-and-whatsapp-with-nlp) ## [Data Science: Customer experience and predictive collection](https://blog.sinnetic.com/en-us/cognitive_services/data-science-customer-experience-and-predictive-collection) Posted by [Gabriel Contreras](https://blog.sinnetic.com/en-us/cognitive_services/author/gabriel-contreras) | Jun 23, 2021 2:53:08 PM According to our Consumer Pulse observatory, 63% of interviewees report a reduction in their... [ CONTINUE READING ](https://blog.sinnetic.com/en-us/cognitive_services/data-science-customer-experience-and-predictive-collection) ### Grow your business more quickly with our solutions for inbound and ecommerce Discover a new digital growth model that attracts visitors, converts them into leads, transforms them into customers and then follows them in the after-sales phase, managed with traditional trade methods or with ecommerce. TAKE AN APPOINTMENT [![WhatsApp](https://upload.wikimedia.org/wikipedia/commons/6/6b/WhatsApp.svg) ](https://wa.me/573180242831) ```json { "@context" : "http://schema.org", "@type" : "BlogPosting", "articleBody" : "As organizations open new client interaction channels, unstructured information capture grows (text, voice, image, etc.). Most chat systems allow for parametrization of instruction sequences but do not inform the business about the conversation or narrative within the chat. This blog seeks to expose some ideas from our consulting experience to make omnichannel an inexhaustible source of insights for the business. The table below shows some questions our clients have asked with relation to the chat’s and WhatsApp's information; it also shows what has been done with the answers Usage case Business questions Actions are taken after obtaining an answer User Experience What is the sequence of conversation? How does it begin, mature and end? What is the client’s emotionality throughout the conversation? What triggers positive or negative emotions? Improve funnels and bottlenecks in the bot’s instruction tree. Detect negative emotional states in advance to accelerate short answers or transition from machine to a human agent. Isolate information to feed models and predict client churn. Audits What are the main needs the user seeks to solve during the chat? Does the agent respond appropriately? (If a human agent) Does the bot have a parameterized response? (if an artificial agent) Customer service clinic. Enable new functions in the channel. Expedite routes and create quick access to specific needs. Improve channel positioning and performance Innovation What user questions are not parameterized in the bot? Unresolved needs? Develop new functions, products or services within the chat. When looking at them closely, the chat’s information has several complexities: A sequence defined by questions and answers very similar to a Ping-Pong game. Short answers that resemble a telegram or tweet. Excessive courtesy, specifically from the agent, makes the conversation’s mood hard to detect. Our consulting process is based on integrating linguistic methodologies to artificial intelligence as shown below: Data integration: Creation of a text datamart that integrates the chats in a sequence, which makes it easier to access enriched text. A text datamart disaggregates the various orthographical forms in a word into tokens. People often do not chat with proper grammar and the absence of an accent or changing an s to a c can change the meaning of a word. Sample design for training: Sample conversation sets are required to train the model. Given that channels evolve and add functionality over time and the client needs change, the sample design must include conversations from different months, moments during the day, days of the week, etc. This also helps to consider the possible seasonal effects that may impact the model’s narrative context. Text mining: Based on the chat set, we carry out parsing and filtering to isolate keywords that appear frequently and are associated between them. Synonyms are cleaned and a list of keywords associated with the business is created so signal detection is honed. This allows extracting core themes. Creation of taxonomies: Using linguistic heuristics and methodologies, we crate thematic taxonomies associated with the needs of the business. This helps classify client questions, products, requirements, services, agent responses and others. Association rules: After classifying the text into key taxonomies for the business accordingly, we analyze the conversation sequence: how it begins and how it ends. With this in mind, rules of association help identify frequent sequences with a high probability of occurrence, which allows to segment conversations according to narrative sequences and create interpretations based on the journey. Context analysis: Using boolean rules and advanced linguistics, taxonomies become automatic classifiers based on context, helping to determine feelings, irony, sarcasm and reproach to finish with a complete textual analysis. We have had the opportunity to support various organizations in their unstructured data analytics challenges for chats. It will be our pleasure to share these experiences with you, let’s meet. South America Mexico and Central America", "author" : { "@type" : "Person", "name" : "Gabriel Contreras", "sameAs" : "", "url" : "https://blog.sinnetic.com/en-us/cognitive_services/author/gabriel-contreras" }, "dateModified" : "23/06/2021", "datePublished" : "23/06/2021", "headline" : "Insight from chatbots and WhatsApp with NLP", "image" : { "@type" : "ImageObject", "height" : 400, "url" : "https://fs.hubspotusercontent00.net/hubfs/8862358/NLP%20ingles%20715x374.jpg", "width" : 750 }, "mainEntityOfPage" : "https://blog.sinnetic.com/en-us/cognitive_services/insight-from-chatbots-and-whatsapp-with-nlp", "publisher" : { "@type" : "Organization", "logo" : { "@type" : "ImageObject", "url" : "https://f.hubspotusercontent00.net/hubfs/8862358/web/logo-sinnetic--slogan-web.png" }, "name" : "SINNETIC LATAM", "url" : "blog.sinnetic.com" } } ``` ```json { "@context" : "http://schema.org", "@type" : "BlogPosting", "articleBody" : "According to our Consumer Pulse observatory, 63% of interviewees report a reduction in their savings capacity and 34% register increased debt levels for two consecutive years. This last portion reports a significant reduction of perceived well-being of 21%. This indicates that the economic tension resulting from the pandemic could be posing challenges of debt payment capacity in consumers; the payment delinquency phenomenon extends to commercial loans. Collection is a key contact point that is rarely shown in the Customer Journey Map. Depending on how this is done, it could generate problems for the customer experience. The purpose of this blog is to analyze best practices with respect to “Predictive Collection” and suggest some ideas that stem from our consulting experience with BPO companies (Business Process Outsourcing), banks and insurance companies. BEST ANALYTICS AND DIGITAL PRACTICES IN COLLECTION PROCESSES Analytics applied to collection processes could be represented by a metaphor of a triangle with three vertices. Collection function Generation of process data and results with analytical purpose. Data-centric Technology favoring push and pull strategies during the process. Operational excellence Integrating operations and optimization research to the collection process. The Collection function Characteristic features that indicate trends from the following perspective: Creation of 7 * 24 self-managed channels to capture customer data. From this perspective, the customer can auction payment options according to his particular possibilities. Auction and price optimization algorithms would be scheduled for these self-managed platforms. These algorithms take advantage of price sensitivity analytics to estimate the optimal level of debt reduction needed to motivate the payment without “leaving money on the table”. MaR (Money at Risk) and PtP (Propensity to Pay) models to support collection decisions and trigger user campaigns. Behavioral sensitive: Automatized collection actions, as well as the models, must be aimed at the behavioral changes of the user. Transactional systems would have to capture these behaviors using monitoring technologies and establishing the user’s Empirical Journey. Next Best Action: In accordance with the price elasticity (Minimum reduction of the obligation needed to raise the PtP (Propensity to pay) models must be programmed in order to propose actions to the user, motivating their payment behavior. In this sense, advances in behavioral economics with particular emphasis on Nudge would support the process motivationally. Operational excellence. A common denominator in collection processes is that they excel at storing results (collection), but are not very efficient in storing process data; generally, the process lacks variables such as: Time between processes Notes from the collection analyst Information from the “People analytics” agent Omnichannel information - chat, bots and WhatsApp- Generally, the process is optimized rather than the result; without data thereof, it is impossible to achieve optimization. In this sense, an essential part of creating a “data-centric” pro “Customer Experience” collection model is to collect process information consistently. As a result of our consulting experience with BPO companies, banks, insurance companies and other companies in the financial sector, we have built a roadmap of the collection process’ analytic evolution, avoiding affecting the user experience as shown below: The operation and self-managed data are captured in a data model we call COLLECTION SIGNAL REPOSITORY. A governance and data quality layer is deployed over this data model to feed five core task nuclei: Collection signals analytics Triggering collection campaigns Feedback and collection Optimization and management Reporting We believe we can add value to your collection strategy by integrating market research, data analytics and artificial intelligence. We will contact you according to your availability South America Mexico and Central America", "author" : { "@type" : "Person", "name" : "Gabriel Contreras", "sameAs" : "", "url" : "https://blog.sinnetic.com/en-us/cognitive_services/author/gabriel-contreras" }, "dateModified" : "23/06/2021", "datePublished" : "23/06/2021", "headline" : "Data Science: Customer experience and predictive collection", "image" : { "@type" : "ImageObject", "height" : 400, "url" : "https://fs.hubspotusercontent00.net/hubfs/8862358/Cobranza%20predictiva%20ingles%20715x374.jpg", "width" : 750 }, "mainEntityOfPage" : "https://blog.sinnetic.com/en-us/cognitive_services/data-science-customer-experience-and-predictive-collection", "publisher" : { "@type" : "Organization", "logo" : { "@type" : "ImageObject", "url" : "https://f.hubspotusercontent00.net/hubfs/8862358/web/logo-sinnetic--slogan-web.png" }, "name" : "SINNETIC LATAM", "url" : "blog.sinnetic.com" } } ```