<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Clevergy - Medium]]></title>
        <description><![CDATA[Tech, product and business posts explaining the clever.gy journey and the tech approaches we are following. - Medium]]></description>
        <link>https://medium.com/clevergy?source=rss----630073e22042---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Clevergy - Medium</title>
            <link>https://medium.com/clevergy?source=rss----630073e22042---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 00:33:54 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/clevergy" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Entendiendo al usuario con Amplitude]]></title>
            <link>https://medium.com/clevergy/entendiendo-al-usuario-con-amplitude-32500ebe53e5?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/32500ebe53e5</guid>
            <dc:creator><![CDATA[Alvaro Perez Bello]]></dc:creator>
            <pubDate>Wed, 25 Oct 2023 11:49:52 GMT</pubDate>
            <atom:updated>2023-10-25T11:49:51.971Z</atom:updated>
            <content:encoded><![CDATA[<p>En <a href="https://www.clever.gy/">Clevergy</a> tenemos una cultura de datos muy fuerte. Desde el primer día, tanto el equipo de ingeniería como el de producto hemos considerado fundamental, instrumentar el producto y los sistemas para poder tomar mejores decisiones. Nuestra estrategia de datos se basa en dos patas principales: trazabilidad de los eventos de dominio e independencia en las consultas.En producto, registramos todos los eventos de dominio en BigQuery. Esto significa que todos los eventos de entidades de dominio de la plataforma quedan guardados en una base de datos. Gracias a este proceso, podemos revisar y auditar, todos los sucesos que ocurren dentro de los sistemas de Clevergy. Si tienes curiosidad, puedes leer sobre la parte técnica en <a href="https://medium.com/clevergy/real-time-analytics-from-event-sourcing-dcfe5ac67c0a"><strong>este blog</strong></a>.</p><p>Además de guardar todos estos datos en BigQuery, hacemos que cada persona sea responsable de hacer sus propias consultas a BigQuery. Aunque aprender SQL tiene una curva de aprendizaje alta, todos los datos están al alcance de cualquier persona del equipo, y gracias a ello:</p><ul><li>Tenemos registrados todos los eventos de la aplicación, por lo que podemos auditar cualquier flujo de usuario</li><li>Podemos entender el histórico de una métrica sin necesidad de pedirle a ingeniería que instrumente ese nuevo flujo</li><li>Hay una fiabilidad del 100% entre los datos que analizamos y los datos que tenemos en las bases de datos operacionales.</li><li>Está desacoplado de producción, una consulta que analice todo nuestro histórico, no afectará al rendimiento de producción.</li></ul><h3>La flexibilidad tiene un precio</h3><p>Si estamos tan contentos con BigQuery, ¿por qué nos hemos planteado utilizar Amplitude?</p><p>BigQuery en sí no es una herramienta de analítica enfocada en el usuario. Es una base de datos que gestiona muy bien grandes volúmenes de información. En BigQuery, hacer consultas analíticas para entender cómo se está comportando la plataforma es fácil. Lo que no es fácil es relacionar eventos de la plataforma con comportamientos del usuario.</p><p>Podemos inferir un inicio de sesión en base a las peticiones de login y de carga de los módulos. Pero una vez cargada la información, ¿cuánto tiempo están? ¿5 minutos? ¿Qué pasó con las personas que se registraron la semana pasada en comparación con el mes pasado?</p><p>Cuando quieres hacer preguntas sobre el comportamiento del usuario, las consultas se van haciendo más complicadas.</p><p>Para ver la complejidad a la que me refiero, aquí podéis ver un análisis por cohortes que realizamos de los usuarios que se registran en la aplicación. Para poder hacer esta gráfica, el equipo de producto tuvo que invertir aproximadamente 8 horas y aun así no fuimos capaces de hacer que la consulta fuera flexible para filtrar una función del cliente.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*YG_iBGhcViA_2SXV.png" /></figure><p>Esto tiene las siguientes implicaciones:</p><ul><li>Dedicamos tiempo extra en algo que no aporta valor en sí (realizar una consulta para ver datos)</li><li>Tiene una curva de aprendizaje alta, y no todo el equipo es capaz de evolucionar este tipo de consultas.</li><li>Barrera alta para utilizar los datos. Conociendo el esfuerzo que llevaba sacar algunas conclusiones, al final no lo medíamos porque no era algo “prioritario” invertir ese tiempo en calcularlo.</li></ul><h3>Amplitude, el mejor complemento para BigQuery</h3><p>Desde que nos dimos cuenta de que necesitábamos un refuerzo, empezamos a mirar diferentes opciones, y en concreto, <a href="https://amplitude.com/startups">el programa para startups de Amplitude</a>.</p><p>Amplitude es una plataforma de analítica que nos permite asociar el comportamiento que tienen los usuarios en nuestra aplicación de una manera sencilla, con gráficas y análisis colaborativos, la mayoría ya creados.</p><p>Amplitude es una de las herramientas más reconocidas en el ámbito de analítica de datos para productos. Aun así, antes de implementarla por completo verificamos dos cosas que para nosotros eran esenciales:</p><ul><li>La primera, comprobar que resolvería el principal problema que teníamos: agilidad a la hora de responder una pregunta a nivel de negocio sobre cómo se comporta el usuario. Desde el tiempo de implementación por parte de desarrollo, hasta la creación del dashboard por cualquier persona del equipo.</li><li>La segunda, confianza en el dato. Queríamos poder trazar las métricas y contrastarlas con la información de BigQuery para poder crecer con ambos sistemas. Para ello, analizamos que hubiese menos de un 1% de desviación en algunas métricas básicas como usuarios registrados y usuarios activos.</li></ul><p>Para hacer todas estas pruebas, nos dimos de alta con el plan Starter ya que era suficiente para empezar a trabajar con la herramienta y verificar estos requisitos.</p><h3>Primeros pasos en Amplitude</h3><p>El equipo de Amplitude estuvo pendiente durante la implementación y nos ayudó a enfocar bien las pruebas. Para acometerlas, decidimos no captar toda la información por defecto que recoge el SDK de Amplitude, ya que genera muchos eventos que no eran relevantes para nuestros casos de uso.</p><p>En este sentido, Amplitude tiene una filosofía a la hora de hacer analítica distinta a la nuestra con BigQuery. Cuanto menos monitorices mejor. No hay una mejor que la otra, si guardas todo, es fácil realizar análisis nuevos y no tienes que esperar a captar datos. Por otro lado, monitorizar sólo los casos de uso que vas a usar hace que tengas menos entidades en la plataforma, haciendo el proceso de búsqueda mucho más sencillo.</p><p>Lo primero fue decidir qué casos de uso íbamos a medir, y trazar eventos específicos para ello. Así fue como, tras una semana definiendo los flujos y otra implementando las trazas en la herramienta, empezamos a tener los primeros datos.</p><p>La primera comprobación que hicimos fue intentar realizar el mismo análisis de cohortes de usuarios que teníamos en BigQuery y Looker Studio. Parecía magia, gracias a los análisis preparados de Amplitude tardamos dos minutos en hacerlo.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*fZm4WOc2TW1ZPKiM.png" /></figure><p>También hicimos algunas validaciones con éxito para comprobar la diferencia entre usuarios y sesiones entre ambas plataformas.</p><p>Es tentador querer implementar una herramienta así. Parece que, gracias a ella, todas las respuestas a tus preguntas aparecerán por arte de magia en la herramienta.</p><p>No obstante, la realidad es que el éxito no está en las respuestas, sino en las preguntas. Saber cuántos usuarios se meten todos los días en la aplicación, no te llevará muy lejos. Sin embargo, entender, qué patrones hacen que un usuario vuelva a la aplicación, sí.</p><h3>Caso práctico: mejorando el funnel de onboarding</h3><p>Uno de los proyectos que hemos realizado con Amplitude desde que empezamos ha sido optimizar el funnel de onboarding.</p><p>En Clevergy, necesitamos datos personales de cada usuario para tener acceso a los datos de consumo de su casa, además de otros datos como (personas que viven en casa, metros cuadrados, electrodomésticos…) con el objetivo de ofrecerle el mayor partido a la aplicación.</p><p>Esto supone un gran reto en el onboarding, cuantos más datos pedimos, mayor es el valor que aportamos al usuario, pero menor es la conversión.</p><p>Hasta la fecha, este flujo de onboarding lo habíamos analizado teniendo como objetivo el acceso a los datos del usuario. Con este objetivo en mente, hemos conseguido en los últimos meses aumentar la conversión de un 30% a un 78%.</p><p>Pero ¿cuántos de esos usuarios vuelven a la aplicación en un rango de 24 horas después de que tengamos acceso a sus datos? Decidimos medir el flujo en Amplitude y a la semana, empezamos a observar las primeras respuestas.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*3cHunIFaZy2BlkHd.png" /></figure><blockquote><em>¡Un 5%!</em></blockquote><p>Nos habíamos enfocado tanto en mejorar la conversión del acceso a datos, que se nos había olvidado que la conversión que realmente importa es mostrarle esos datos al usuario.</p><p>Las buenas noticias son, que este resultado era esperado, ya que habíamos visto proporciones similares entre los usuarios que completaban su registro y aquellos que volvían a la aplicación.</p><p>La primera idea que se nos ocurrió parecía evidente: reducir el tiempo de acceso a los datos del cliente. No podemos conseguir que los usuarios accedan a la aplicación en menos de 24 horas, si no somos capaces de acceder a esos datos en ese tiempo. Este comportamiento es por un motivo técnico. No podemos dar de alta a un usuario mediante un servicio API. Por lo tanto, dar de alta al usuario en tiempo real quedaba descartado. Al final encontramos una solución poco convencional, usando distintas herramientas. Después de probar con varias soluciones hemos sido capaces de reducir ese tiempo de acceso a 3 minutos de media.</p><p>Desgraciadamente, al implementar este cambio comprobamos que los usuarios no volvían… Para cuando teníamos acceso a los datos, el usuario había cerrado la aplicación y se había olvidado de nosotros.</p><p>¿Qué podíamos hacer para que ese usuario no se fuera?</p><p>Alargar el onboarding. Mantener la atención del usuario dentro de la aplicación para que nos diera tiempo a acceder a sus datos y poder mostrarle su información.</p><p>Esas pantallas cumplían dos funciones:</p><ul><li>Explicarle al usuario que esperar antes de utilizar la aplicación para acelerar su proceso de onboarding</li><li>Solicitarle información al usuario para personalizar su experiencia</li></ul><p>Es importante destacar, que ya habíamos comprobado que la tasa de usuarios activos aumentaba de un 17% a un 57% en usuarios que completaban información como el perfil de su casa dentro de la aplicación, por lo que incluir este paso en el onboarding parecía obvio.</p><h3>Resultados</h3><p>La conversión al terminar de validar este nuevo flujo aumentó al <strong>24,7%.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*FltFfW6E28HEXKST.png" /></figure><p>Aunque esto es solo el principio y la muestra no sea tan grande, la idea de este artículo no es mostrar cuánto ha mejorado el número, sino contar abiertamente cómo Amplitude nos ha ayudado durante todo este proceso de reflexión y pruebas.</p><p>Además, no solo hemos podido obtener muestras absolutas del funnel, sino que además obtenemos otro tipo de información valiosa que nos permite realizar nuevas hipótesis, como por ejemplo:</p><p>El tiempo que tardaba el usuario en completar el perfil con los datos de su casa.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*OnSlcAWfizLaEwdd.png" /></figure><p>O la posibilidad de crear cohortes y medir la diferencia de retención entre los usuarios que se registraron con el nuevo onboarding y el antiguo.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*_BFPEb-5jZIwB9bp.png" /></figure><p>Sin duda, esta facilidad y agilidad para de responder a preguntas sobre cómo se comportan los usuarios ha supuesto un gran avance en la cultura de datos de Clevergy.</p><p>El impacto con el rediseño de la aplicación ha sido enorme. No solo por haber aumentado la conversión entre registro y primera sesión útil del usuario sino por dos aprendizajes que no habríamos sido capaces de medir sin Amplitude:</p><ul><li>Aumentando el proceso de onboarding, disminuimos el porcentaje de registros completados, pero las personas que terminaban el registro estaban más preparadas para usar la aplicación. Esto tuvo como consecuencia un aumento en el número neto de personas que acaban visitando la aplicación.</li><li>Cuando solo analizas un funnel a través de trazas y te abstraes del proceso que realiza el usuario en tu aplicación, tiendes a optimizar métricas que no son importantes, como reducir el tiempo de registro o aumentar el número de registros completados.</li></ul><h3>Claves que hemos aprendido durante el camino</h3><p>Durante el camino de implementación de Amplitude hemos encontrado aspectos positivos y negativos. Por eso, antes de terminar, quiero compartir algunas lecciones que hemos aprendido en el proceso.</p><h3>Definir un caso de uso en vez de medir todo</h3><p>En el pasado he probado varias herramientas de analítica, por lo que tengo claro que cuando tienes acceso a todo lo que ocurre en tu aplicación, se genera mucho ruido que dificulta encontrar lo que quieres.</p><p>Con esta premisa decidimos probar Amplitude sin medir todo. Pensamos qué casos de uso concretos queríamos medir y a partir de ahí, trazamos los eventos necesarios desde nuestra aplicación.</p><p>Esto no solo hizo la herramienta más útil, sino que nos dio más confianza en el dato, ya que son eventos que controlábamos al 100%, tanto para crear nuevos como para modificar lo que no cuadraba.</p><h3>Planifica tiempo para generar un histórico</h3><p>Por otro lado, esta virtud nos generó conflicto. Cuando planteas hipótesis necesitas datos, y con Amplitude hemos tenido que esperar para recoger datos.</p><p>Aunque puedes importar datos de varias fuentes, quizá sean casos de uso nuevos, por lo que no esperes usar Amplitude al instante.</p><p>En nuestro caso necesitamos un mes para recoger datos y hacer estos gráficos y comparaciones.</p><h3>Involucra al equipo de desarrollo para una correcta implementación</h3><p>Tanto al principio de la implementación, como para definir nuevos eventos o depurar alguna métrica que no se esté midiendo bien, hemos necesitado una ayuda continua del equipo de desarrollo.</p><p>En nuestro caso, las ventajas de Amplitude están por delante a las desventajas, pero sí que hay que tener en cuenta que vas a necesitar recursos técnicos para medir bien. Esto implicar dedicar tiempo a definir requisitos, planificar el desarrollo y esperar un tiempo para recoger datos que te permitan decidir.</p><h3>La curva de aprendizaje por parte de producto es muy buena</h3><p>A diferencia de otras herramientas, Amplitude ha sido muy sencillo de aprender a utilizar. Además, cuenta con multitud de plantillas que te ayudan a saber cómo puedes sacarle el máximo potencial a la herramienta.</p><p>Esto fue un gran punto a favor para nuestra empresa, ya que cuando usábamos BigQuery y Looker Studio, muchos de los gráficos eran complejos a nivel de SQL, y no todas las personas del equipo están capacitadas para realizarlos o entenderlos.</p><p>Ahora cualquier persona del equipo puede entrar en Amplitude y crear su propio análisis.</p><h3>Próximos pasos en Clevergy</h3><p>En este artículo hemos mostrado cómo Amplitude nos ha ayudado a mejorar el funnel de onboarding de nuestra aplicación, y cómo esta herramienta ha impulsado nuestra cultura de datos y nuestro aprendizaje.</p><p>Creemos que Amplitude es una herramienta muy útil para cualquier empresa que quiera basar sus decisiones en datos, y por eso recomendamos su uso a otros equipos del sector. Además, nos gustaría seguir explorando otras funcionalidades de Amplitude que nos permitan medir otros aspectos de nuestra aplicación, como la satisfacción del usuario, el engagement o la fidelización.</p><p>Amplitude nos ha permitido obtener los datos de comportamiento de nuestros usuarios con más agilidad, y mejorar nuestra aplicación y su experiencia de uso. Lo que más me ha gustado de este proceso es que nos ha motivado a basar todas nuestras hipótesis y épicas de producto en datos. Ahora nadie tiene excusa para no entrar en Amplitude, sacar sus conclusiones y compartirlas con el equipo.</p><p>Esperamos que este artículo te haya resultado interesante y útil, y que te anime a medir más en tu propio producto.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=32500ebe53e5" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/entendiendo-al-usuario-con-amplitude-32500ebe53e5">Entendiendo al usuario con Amplitude</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Real-time analytics from event sourcing]]></title>
            <link>https://medium.com/clevergy/real-time-analytics-from-event-sourcing-dcfe5ac67c0a?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/dcfe5ac67c0a</guid>
            <category><![CDATA[clevergy]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[distributed-systems]]></category>
            <category><![CDATA[event-sourcing]]></category>
            <category><![CDATA[data-mesh]]></category>
            <dc:creator><![CDATA[Javier Martinez Valbuena]]></dc:creator>
            <pubDate>Tue, 25 Apr 2023 10:24:14 GMT</pubDate>
            <atom:updated>2023-04-25T11:27:08.860Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kC5DANksLrdwd3WP.jpg" /><figcaption>They are not loose pieces, they are meaningful routes. The importance of data governance.</figcaption></figure><blockquote>This post was written alongside <a href="https://medium.com/u/d2ec4cae67ff">Juan López López</a></blockquote><p>At <a href="https://clever.gy/">Clevergy</a>, we have always kept a strategy: to ensure that <strong>any event on the platform can provide raw and processed real-time information that can be visualized, transformed, and managed by every person within the company (tech, business, product, etc.)</strong>. This strategy also includes the will to <strong>avoid any kind of development</strong> required for data availability, delegating the transport and persistence work to the data platform, and avoiding also heavy classical ETL processes that require knowledge of third-party applications.</p><p>In many companies, there is a data team processing information, but often this team lacks enough context about data changes and is prone to making mistakes. To solve this problem and to facilitate the flow of information among all actors in the company, at Clevergy we have decided to adopt the strategy of data democratization.</p><blockquote>Through a model based on the idea of <a href="https://martinfowler.com/articles/data-mesh-principles.html">Data Mesh</a>, each domain exports every event of creation or modification of <strong>domain entities</strong> from the operational plane (operational services in the application) to the analytical plane, making all information, raw, from the platform available in the data warehouse.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ue8p6_C-quxc_ies.png" /><figcaption><em>Figure 12: Data mesh architecture from 30,000 foot view. Image from </em><a href="https://martinfowler.com/articles/data-monolith-to-mesh.html">https://martinfowler.com/articles/data-monolith-to-mesh.html</a></figcaption></figure><h4>Our short but intense story about domain events</h4><p>We have always focused on agile development from the very beginning of the platform design process, allowing us to have business functionalities as quickly as possible, trading off the technical debt generated to test the features with the users. We have always asked ourselves when we should refactor each piece of software, taking into account the growth and projection needs of the platform. Regarding event design, <strong>the application started with an entity-action event design, which we recently decided to refactor into an entity-based design model with the operation field.</strong></p><p>Event-based architectures can become unwieldy and difficult to manage when dealing with large amounts of data or complex systems. <strong>By switching to an entity-based design</strong>, it may be <strong>easier to organize and analyze the data</strong>, implement changes and updates, <strong>and generate insights and analytics</strong>. In addition, <strong>entities can be separated for security reasons</strong> and published to different topics, allowing for modularity and reuse of components across the system. <strong>High-throughput events can also be published to dedicated topics for optimal performance.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*op7wJ3uE19XtsL3neRTAIw.png" /><figcaption>Now we have a topic by domain entity with Protobuf schemas</figcaption></figure><p><strong>Regarding the analysis point, we have always tried to dump every single event on the platform into an analytics database, allowing the history of each business entity to be available.</strong> With the current entity-based topic design, we have naturally managed to group information by domain entities, making it more easily accessible.</p><p>The domain-specific information is preserved across the entire information channel, spanning from the operational to the analytical domain. <strong>However, the system must remain resilient to changes, and because of that, we have implemented a strategy that involves using </strong><a href="https://protobuf.dev/"><strong>Protobuf</strong></a><strong> schemas as a contract for each event</strong>. This allows us to version the models and decouple the infrastructure layers.</p><h4>Data-informed decision making</h4><p>A <a href="https://en.wikipedia.org/wiki/Data-informed_decision-making">data-informed</a> approach <strong>enables organizations to make informed decisions based on data insights</strong>, rather than relying on intuition or assumptions, or even only on data. We use data to understand user behavior, improve user experience, optimize performance, and create better product offerings. To summarize, Clevergy <strong>manages data as a critical input for decision-making and development.</strong></p><p><strong>With each change our domain entities have, an event is launched to the platform and dumped into an analytics database. This event finishes in a table, organized by domain entities, in our data warehouse. With that, we have created real-time dashboards for every important business metric.</strong> User conversion, application usage metrics, module usage metrics, user usage patterns, and more have been turned into information boards by business users themselves, allowing us to measure the impact of changes and establish guidelines for future functionality development.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/0*EfINWuZyGhz6PSZV" /><figcaption>Data-driven vs data-informed strategy</figcaption></figure><h4>Free IT Crowd</h4><p>Data is a valuable asset that can provide critical insights into customers, markets, and operations. However, analyzing and interpreting data can be a time-consuming and complex task, which can be a challenge for small businesses with limited resources. <strong>Our solution to this is to empower the whole organization to do the data analysis by learning SQL. </strong>With that, tech staff, always busy, is not a blocker to that anymore <strong>improving the overall efficiency of the organization.</strong></p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fgiphy.com%2Fembed%2FdbtDDSvWErdf2%2Ftwitter%2Fiframe&amp;display_name=Giphy&amp;url=https%3A%2F%2Fmedia.giphy.com%2Fmedia%2FdbtDDSvWErdf2%2Fgiphy.gif&amp;image=https%3A%2F%2Fi.giphy.com%2Fmedia%2FdbtDDSvWErdf2%2Fgiphy.gif&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=giphy" width="435" height="236" frameborder="0" scrolling="no"><a href="https://medium.com/media/aa46dec20a0d433201be0a52abb24235/href">https://medium.com/media/aa46dec20a0d433201be0a52abb24235/href</a></iframe><p>One of the primary advantages of giving the autonomy of data analysis to product staff is the ability to leverage their domain expertise. <strong>Business and product people</strong> are often the ones who understand the data best, as they <strong>have direct experience with users, markets, and operations</strong>. By empowering them to analyze the data, businesses can gain more accurate and relevant insights that can drive informed decision-making.</p><p><strong>When everybody is able to analyze data, they are more likely to share their insights and collaborate with colleagues across departments.</strong> This can lead to new ideas and innovations that can drive business growth and competitiveness.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Q13gm4s1z_UPlBzoqJmRKA.jpeg" /><figcaption>Looker dashboard sample</figcaption></figure><h4>Real-time analytics as part of our observability strategy</h4><p>Observability refers to the ability to measure, monitor, and understand the internal state of a system or application based on its external outputs or behaviors. Not only an exception or an error in our code shows a problem in our system, but if we are not having new users or no users are doing transactions at a specific time, probably your platform has to experiment with some problems.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*tSZ451BzFNOFyk9L.png" /><figcaption>An Instana slide to explain what observability is.</figcaption></figure><p>For us, the value of having real-time data is not to have real-time analysis of the features (they need some time to understand if they work well with the users). <strong>The value of real-time data is to use this data to understand if our system is working properly.</strong></p><p>The next step here is to build an alert system to inform us when we have this lack of business events as we have with code errors. Now we are checking it manually.</p><h4>GCP as a key part of the strategy</h4><p>Google Cloud Platform (GCP) is a cloud computing service that provides a wide range of tools for building, deploying, and managing applications and services. Two of the most valuable services offered by GCP are Pub/Sub and BigQuery, which allow users to store, process, and analyze large amounts of data in real time. <a href="https://cloud.google.com/blog/products/data-analytics/pub-sub-launches-direct-path-to-bigquery-for-streaming-analytics">The combination of these services</a> can provide significant advantages for storing and analyzing event data produced by changes in the domain entities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/570/1*JWeJ34NBCid6D2ucgIua0w.png" /><figcaption>Pub/sub + Protobuf + BigQuery</figcaption></figure><p><strong>All good and agile data strategy relies on the possibility of using a data platform that facilitates work and avoids extra developments</strong>. In our case, we use Google’s cloud infrastructure to deploy services and connect Pub/Sub topics to an analytics database in BigQuery via subscribers. We use model versioning for the structure of the topics and link the event schema to the BigQuery tables. All updates to any domain model are available in the table with the same name, and we maintain the nomenclature in both the topics and tables to facilitate maintenance.</p><h4>Conclusions</h4><p>At Clevergy, we have sought and tested a way to obtain data from the platform that allows us to make business and development decisions.</p><p>To summarize the main ideas we have talked about:</p><ul><li><strong>Entity-based design models </strong>with a topic by entity have better scalability.</li><li><strong>Data-informed</strong> <strong>decision strategy </strong>is perfect to build a product in an agile way, but, obviously, you need the data. And, ideally, democratized.</li><li><strong>SQL is an incredible weapon</strong> to empower everybody to understand how the software is working and to think of a good product strategy.</li><li><strong>Real-time</strong> <strong>data</strong> is a key part of the <strong>observability</strong> strategy.</li><li><strong>GCP </strong>has an incredible suite to work with data.</li></ul><p>We have tested this strategy over the last year and we are proud to be able to share with everyone this line of work that allows for improved performance and work speed for similar projects that want to adopt it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=dcfe5ac67c0a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/real-time-analytics-from-event-sourcing-dcfe5ac67c0a">Real-time analytics from event sourcing</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Clustering houses by energetic profiles]]></title>
            <link>https://medium.com/clevergy/clustering-houses-by-energetic-profiles-6192096e5882?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/6192096e5882</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[software-engineering]]></category>
            <category><![CDATA[statistical-analysis]]></category>
            <category><![CDATA[energy-efficiency]]></category>
            <dc:creator><![CDATA[Javier Martinez Valbuena]]></dc:creator>
            <pubDate>Thu, 09 Mar 2023 15:09:49 GMT</pubDate>
            <atom:updated>2023-03-13T09:37:13.280Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*S0I5t82jz4gTeqPn" /><figcaption>Photo by <a href="https://unsplash.com/@brenoassis">@brenoassis</a>, unsplash.com</figcaption></figure><p>In <a href="https://app.clever.gy/">Clevergy</a> we offer users several functionalities to <strong>save energy</strong>. We show users how they consume at home and how to have a better relationship with energy. We help them giving energy insights such as the consumption by disaggregation of appliances, the hourly consumption by time ranges, the estimated consumption in bill cost, the possibilities of saving through the modification of contracted power or a change of electricity supplier and many more energy recommendations.</p><h3>The problem</h3><p><strong>At this time, we are going to refine a functionality that allows an understanding of how much they are spending compared with “similar”households to yours</strong>. 120 kWh is not too much or little, it is what it is, and it is essential to understand if you can improve it. To solve this problem we have faced the challenge of classifying the homes of our customers in similar groups in terms of the energy profile of the houses. Once the groups where made we placed them in a percentile of expenditure compared to other houses allowing the user to understand how well they are spending.</p><p>The first thing we have to define is: <strong>What does it mean “similar homes”?</strong></p><p>In this case, we approach the question from a multifactorial point of view.</p><p>Similar dwellings cannot be dwellings that have similar consumption. A small house that consumes n kWh in a cold place basically due to a basal consumption of heating, has nothing to do with a large house that consumes the same energy at peak hours due to the consumption of kitchen appliances.</p><p>On the other hand, we cannot classify by spatial “neighbors” either, since very close houses may have nothing to do with each other from the point of view of energy consumption.</p><h3>Data pre-processing</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*R-l3nenlfh4tDUOu.png" /></figure><p>We have studied comparable factors when classifying houses and have found several key elements to be able to group them.</p><ul><li><strong>Square meters</strong>. From an energy point of view, it is practical to classify dwellings primarily by size.</li><li><strong>Electricity consumption potential</strong>. We have seen that no matter how similar the houses are in terms of characteristics, it does not make sense for example to group as similar a house with gas heating and one with electric heating (statistical studies indicate that electric heating is almost 50% of household consumption), so we propose to estimate a volume of Kw “available” by multiplying the devices that we know that a house has by its average nominal power.</li><li><strong>Climate profile</strong>. Spain is a country in which we can find very different climatic profiles even within the same regional organizations. We have used for our grouping the Mediterranean, cold inland, warm inland, Atlantic, and Cantabrian profiles. We have assigned each complete province to one of these climates.</li><li><strong>Housing type</strong>. We know that this is an assumption that can have nuances, but we prefer to use it to characterize the level of energy performance we take as distinct single-family homes and apartments.</li></ul><p>For this first version we are using only those few features, but we are currently working on really amazing features to enrich our houses dataset, such as real-time climate profile, state information about housing properties, energetic profile (isolation level), and satellite images to identify roof areas in houses with solar production</p><p>Once we have these columns calculated for each house we proceed to calculate the grouping of dwellings according to these characteristics.</p><h3>A first approach to the solution</h3><p>In the first approach, we proposed the possibility of using Python mathematical libraries to build a service that would perform the classification in three steps:</p><ul><li>A <strong>PCA</strong> <strong>(Principal component analysis). </strong>This filter would modify the dataset dimensions to make it two-dimensional, from two columns to a single pair of values, in order to be able to run K-means against a two-dimensional vector.</li><li><strong>A study of a possible number of centroids</strong>. To achieve an optimal result, we ran a clustering simulation for several values of the total centroids, calculating the error of each run and saving the one with the smallest deviation.</li><li>Cluster assignment to each house using <strong>K-means</strong>.</li></ul><p>Once the cluster IDs are assigned, the houses are persisted again in order to retrieve each set and its consumptions from the services that make up the application. As this process can be heavy it is periodically executed in the background.</p><p>Once we have the classification we use <a href="https://cloud.google.com/vertex-ai">GCP’s VertexAI</a> to train a classification model that allows us to have an API to query in case we receive houses that are not yet grouped, and it returns the cluster most similar to the new house.</p><h3>Easier is better</h3><p>Once we finished the proof of concept of the classifier we realized that the classification process based on the <a href="https://cloud.google.com/vertex-ai">VertexAI</a> service, despite being very comfortable for us, introduces the error inherent to the model added to the error introduced by the K-means algorithm, so we decided to eliminate it from the equation by directly calculating the cluster of a new house through the minimum distance to the centroids.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/710/0*_AAJ2SbjLUwCzP4t.png" /><figcaption>XKCD <a href="https://xkcd.com/1349/"><strong>Shouldn’t Be Hard</strong></a></figcaption></figure><p>On the other hand, and taking into account that our proof of concept does not have a sufficient entity, <a href="https://martinfowler.com/bliki/AnemicDomainModel.html">nor does it fully correspond to any domain</a>, we decided to eliminate the service and include it in one of our services.</p><p>As our services are programmed in Kotlin, we decided to use <a href="https://www.cs.waikato.ac.nz/ml/weka/">WEKA</a>, which also allows us to eliminate the principal component analysis part and do clustering on multiple dimensions.</p><p>This way we eliminate almost all the complexity and have amazing results. And we did it with 4 lines of code!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/880/1*GNOJdDQw7Q72Uz3eh2wEDw.png" /></figure><h3>Conclusions</h3><p>In an unsupervised analytical process, as in the case of the application of the K-means algorithm, <strong>it is essential to make iterations that allow us to analyze the quality of the solution</strong>. In this context, we have put effort into preparing a dataset at the input of the algorithm containing processed data in each column to give us an ideal result.</p><p>We always use control metrics to know if we gain precision in our systems, and in this specific use case, we work with a baseline that allows us to know if we should optimize the number of clusters based on the accumulated errors in each iteration.</p><p>In engineering, <strong>it is an excellent decision to refactor a process no matter how expensive it has been if the result is going to be lighter and simpler.</strong> It is essential not to be afraid to destroy and redo if there are guarantees that the refactoring will allow easier maintenance and better fault tolerance.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6192096e5882" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/clustering-houses-by-energetic-profiles-6192096e5882">Clustering houses by energetic profiles</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Clevergy´s journey. A startup logbook. Month 5]]></title>
            <link>https://medium.com/clevergy/clevergy-s-journey-a-startup-logbook-month-5-3b8431df4c87?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/3b8431df4c87</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[clevergy]]></category>
            <dc:creator><![CDATA[Juan López López]]></dc:creator>
            <pubDate>Mon, 10 Oct 2022 10:07:40 GMT</pubDate>
            <atom:updated>2022-10-10T10:07:39.908Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*MalbQ8JhYaWmS56T" /><figcaption>Photo by <a href="https://unsplash.com/@samscrim">@samscrim</a>, unsplash.com</figcaption></figure><p>A lot of things happened this month… sometimes it is difficult to assimilate how much you do when you are in startup mode.</p><p>Summarizing we were working on <strong>releasing our mobile apps, growing our B2C product, learning about our errors, Google for Startups, and improving the performance of our platform</strong>:</p><h4>We have <strong>released our </strong><a href="https://apps.apple.com/es/app/clevergy/id1644281726"><strong>iOS</strong></a><strong> and </strong><a href="https://play.google.com/store/apps/details?id=gy.clever.clevergy"><strong>Android</strong></a><strong> apps</strong>.</h4><p>We are really happy about it because anyone can download it now, understand their house consumption, and save money and energy. Two hot takes:</p><ul><li><strong>We were right about our micro-frontends strategy</strong>. This is bringing us a lot of benefits. Not only on the web, with this approach it is very easy to have updated your app. Here is an interesting post about that: <a href="https://ionic.io/resources/articles/micro-frontends-for-mobile-with-ionic-portals"><strong>https://ionic.io/resources/articles/micro-frontends-for-mobile-with-ionic-portals</strong></a></li><li>Uploading<strong> an app to the stores is not easy and it can be frustrating sometimes</strong>. Especially Apple is very picky with its apps and, although their comments help you to improve the app, sometimes they force you to do some changes you think it is not the moment for it.</li></ul><h4>Tech enabling incredible B2C possibilities</h4><p>We <strong>unlock the possibility of having the consumption of every person who </strong>completes <strong>(only with DNI) </strong>the registration in <a href="https://app.clever.gy/">Clevergy</a>. So, if you want to understand your consumption and start saving energy, <a href="http://app.clever.gy/">sign up now for free</a>!</p><h4>New team member´s onboarding.</h4><p><a href="http://www.extremeprogramming.org/">Pair programming and TDD</a> are so good. In the first week, the new member has been uploading small changes directly into production. He brings immense expertise in the domain and the way we work in Clevergy.</p><h4>First postmortem.</h4><p>Shit happens… but errors teach us things. In our case, we had a problem with an iOS security update and some cross-domains in our Auth0 authentication process. We know errors happen, but it is important:</p><ul><li>To show the company what happened.</li><li>To understand what the impact was (it is important to have enough data).</li><li>To take action to avoid the same problem in the future.</li></ul><p>We use the template from Betsy Beyer, Chris Jones, Jennifer Petoff, and Niall Richard Murphy: <a href="https://landing.google.com/sre/book/chapters/postmortem.html">“Site Reliability Engineering”</a>.</p><h4>Threads management</h4><p>When the traffic starts growing and you are managing a lot of third parties requests, you start experiencing real problems. In our case concurrency is under control (🤞) but we did need to work on the threads management. As we use Kotlin, <a href="https://kotlinlang.org/docs/coroutines-overview.html">coroutines</a> help us in non-blocking programming and thread pool management. We recommend you read about it. We will probably write a post about our experience with that later.</p><h4><strong>Queries milliseconds matter</strong></h4><p>We understand that improving the performance of every query in our platform is crucial. Speed is always important, and even more when the traffic starts to grow. We don´t use any ORM or anything similar because we think that every query is important. Moreover, the previous month <a href="https://cloud.google.com/">GCP</a> released a very cool feature, the <a href="https://cloud.google.com/sql/docs/mysql/using-query-insights">Queries insights</a>, that help you to monitor without additional effort your queries.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VJrO7mH3VZbMa72s.png" /></figure><h4>We are in a <a href="https://startup.google.com/">Google for Startup</a>s program!</h4><p>Last but not least, in September we started in the Google Campus of Madrid in the <a href="https://www.campus.co/madrid/growth-academy/">Sustainability program</a>. Very interesting talks, amazing networking, and mentoring, so we thank Google for that. Furthermore, we can use the incredible Google Campus installations to work there.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*VKoamtKOQ1f63lId" /></figure><p>Stay tuned for the next edition!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3b8431df4c87" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/clevergy-s-journey-a-startup-logbook-month-5-3b8431df4c87">Clevergy´s journey. A startup logbook. Month 5</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Clevergy´s journey. A startup logbook. Month 3&4]]></title>
            <link>https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-3-4-321704cc9745?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/321704cc9745</guid>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[clevergy]]></category>
            <dc:creator><![CDATA[Juan López López]]></dc:creator>
            <pubDate>Fri, 02 Sep 2022 10:50:51 GMT</pubDate>
            <atom:updated>2022-09-02T11:21:31.299Z</atom:updated>
            <content:encoded><![CDATA[<h3>The Clevergy´s journey. A startup logbook. Months 3&amp;4</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*tAV3MS0p2gcw_gbz" /><figcaption>Photo by <a href="https://unsplash.com/@mmcgregor">@mmcgregor</a>, unsplash.com</figcaption></figure><p>This time we are going to tell you what we did in these two summer months. Months between vacations, mountains, and beaches. <br>Good work and good news, mainly in these topics: <strong>hiring, b2c, real-time data analytics, and incredible new product features.</strong></p><h4><strong>Tech recruiting is hard.</strong></h4><p>Very hard. You know, <strong>tech is moving the world and there are not enough people to work on that</strong>. You can think that having good perks and conditions brings thousands of CVs but it is not true at all. <strong>You need to work hard in every step of the process.</strong> On the one hand, you need to be clear about the company culture you want to create, and on the other hand, what you expect from the candidates and what you can offer them.</p><p>We put a lot of effort into writing a clear and honest <a href="https://clevergy.notion.site/Clevergy-Mid-Software-Engineer-051646375646437d98712fee6132e8c4">job description</a> and finally we achieve hiring an incredible person that starts this month to work with us. Some <strong>insights</strong> into the process:</p><ul><li>We had the opportunity to interview <strong>incredible people</strong>.</li><li>Explaining <strong>clearly the different steps of the process</strong> is important to people<strong>.</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*V1mimZF8igSUYIKo" /></figure><ul><li>Writing a <strong>good job description </strong>is much more important than you think.</li><li><strong>Pair programming interviews</strong> without bullshit algorithm tests were crucial for us to understand and know the people.</li><li><strong>4 day weeks</strong> are a game changer.</li><li><a href="https://www.getmanfred.com/manfred-daily">Manfred Daily</a> was an incredible speaker to reach good talent.</li><li><strong>Hiring and the work of recruiters is as essential as it is difficult. Hiring good talent builds good teams.</strong></li></ul><h4>Good event ecosystems become real-time data for free (almost)</h4><p><strong>Our platform does new business cases always, so every time a business entity changes in our system </strong>(although nobody read it)<strong>, we are sending an event with the entity to the system</strong> via our <a href="https://cloud.google.com/pubsub">event bus</a>. It enables asynchronous communication between services and follows all the best practices of new architectures.</p><p>But it enables us to do something amazing, to have real-time data with no extra work. <a href="https://cloud.google.com/dataflow/docs/guides/templates/provided-streaming">PubSub has the possibility to save in streaming into Bigquery</a> every event of different topics, so we have a dataset with all of them (a table by event).</p><p>This brings to us a separate data warehouse for analytics and all the information we need to understand how everything is working, analyze the data, and be data-informed.</p><p>Furthermore, we worked to have a clear dashboard where every person in Clevergy can understand how their work is working.</p><h4><strong>B2C</strong></h4><p>We unlocked something incredible. We can access <strong>the consumption of every person in Spain.</strong> This function enables us to have a very interesting <a href="http://app.clever.gy/">B2C product</a> and we are receiving very good feedback about that. Remember that <strong>consumptions and data are always yours.</strong> With Clevergy you can take advantage of that and start saving from today.</p><h4><strong>Product news</strong></h4><ul><li>We are <strong>analyzing the power</strong> you have in your contract and we can <strong>recommend</strong> an important save if it is not fitting to your home habits.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/722/1*8Zlpk-O_Cx9lt0KmUyf1IQ.png" /></figure><ul><li><strong>Rates</strong>. Different rates became different ways to see and understand your consumption. So now you can choose your rate type and with that, we explain to you and offer you different ways to save on your bills.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/754/1*hJ7fSiUlVWWX9_N1muInOA.png" /></figure><ul><li>New <strong>house comparator. </strong>Is 85€ too much? or is it a fair consumption? We want to help you understand your performance according to a comparison between yourself and similar people.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/608/1*BafsgcXm6OqplK2mnQijLQ.png" /></figure><h4>North star metric</h4><blockquote>A North Star metric is the one measurement that’s most predictive of a company’s long-term success.</blockquote><p>We will explain it in another post because it was pretty interesting but finally we decided that our <a href="https://mixpanel.com/blog/north-star-metric/">North Star</a> <a href="https://mixpanel.com/blog/north-star-metric/">metric</a> is the <strong>user savings</strong>. We are building that to reduce the electric consumption in homes so we really think that this is the metric we need to follow if we want to build something useful. Furthermore, <strong>with an incredible product, we are sure we will have a profitable company</strong>.</p><p>Stay tuned for the next edition!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=321704cc9745" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-3-4-321704cc9745">The Clevergy´s journey. A startup logbook. Month 3&amp;4</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Clevergy´s journey. A startup logbook. Month 2]]></title>
            <link>https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-2-d2da0d3fa2b?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/d2da0d3fa2b</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[clevergy]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Juan López López]]></dc:creator>
            <pubDate>Mon, 04 Jul 2022 05:56:52 GMT</pubDate>
            <atom:updated>2022-07-04T05:56:52.150Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*K7x2CJnejaGmPdYc" /><figcaption>Photo by <a href="https://unsplash.com/@xavi_cabrera">@xavi_cabrera</a>, unsplash.com</figcaption></figure><h4>To decide what you are not going to do is everything</h4><p>When you are starting a new project, you have too many ideas in your head, and each one of them looks really brilliant. And maybe they indeed are. However, <strong>tech products take time to build</strong>, especially the development part, so, you <strong>need to</strong> <strong>prioritize and discard some of your t̶h̶o̶u̶s̶a̶n̶d̶s̶ millions of ideas, and that can be frustrating but you have to</strong>.</p><p>In <a href="http://clever.gy/">Clevergy</a>, every two weeks we are doing a product workshop with all the team members to align us to the next priorities we want to do, in the short, medium, and long term. In other words, we are doing a <a href="https://en.wikipedia.org/wiki/Straw_man_proposal"><strong>straw-man proposal</strong></a><strong> of roadmap</strong>. As one will expect, the tasks planned for the long term change a lot as time passes by but, with this exercise, we consolidate our short and medium <strong>goals to achieve our strategy while aligning us</strong>. If we have something clear, it is that:</p><blockquote>Our product is better than yesterday but worse than tomorrow.</blockquote><p>Regarding the things we were working on during this month, here <strong>we highlight the most important</strong>:</p><ul><li>We <a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-1-74c0b589fe8">told you</a> that we were giving <a href="https://www.claap.io/product"><strong>Claap</strong></a> a shot, and it is really useful to be aligned and work in an <strong>async way</strong> (we don´t want to fall into the <strong>meetings spiral</strong>). What is the use we are giving to this tool?</li></ul><pre>When the product team is working on a new prototype/feature, they are creating a Claap to share and ask for feedback/problems from the team. This is creating a lot of engagement with the product and further identify problems in development before starting to develop it. Furthermore, we are doing a weekly summary of new developments/works we are doing to have this documented. Probably we will start to use Claap for more things.</pre><ul><li>Ana started this month as a software engineer in <a href="http://clever.gy/">Clevergy</a> and it has been amazing because she deployed in production in her first week (<strong>a full feature, backend, and frontend</strong>). Pairing is amazing to share knowledge and having automated pipelines and deployments to production environments helps a lot. “<a href="https://apiumacademy.com/blog/evolutionary-architectures-principles-common-characteristics/"><strong>Bring the Pain Forward</strong></a><strong>” </strong>principle of the <a href="https://www.infoq.com/news/2016/03/evolutionary-architectures/">evolutionary architectures</a> is giving us a lot of benefits from the very beginning. If you want to join as Ana did, remember we are hiring! Write <a href="/service/mailto:juan.lopez@clever.gy">me</a> if you are interested.</li><li>We built a <strong>native app</strong>. Maybe it sounds like an incredible work, but the reality is less fancy. Thanks to our microfrontends catalog, we are able to build easily different customized applications. To make it available in a native way, we are using <strong>Flutter and the </strong><a href="https://pub.dev/packages/webview_flutter"><strong>webview plugin</strong></a>. After that, with Flutter, you can export it to iOS and Android with a click. Too easy, and has incredible results, so <strong>I strongly recommend Flutter if you want something simple like us</strong>.</li><li>We are doing <a href="https://www.atlassian.com/continuous-delivery/continuous-integration/trunk-based-development"><strong>trunk-based development</strong></a>. It comes, always, with other practices like very small deployments and other kinds of code reviews (including pair programming). So far, with this approach, we are building new features for <a href="http://app.clever.gy/">our product</a> very fast.</li></ul><figure><img alt="Welcome tab built with five Clevergy microfrontends." src="https://cdn-images-1.medium.com/max/388/1*MOZYKW_YRHNNK0Vktycgiw.png" /><figcaption>Welcome tab built with five Clevergy microfrontends.</figcaption></figure><p>Stay tuned for the next edition!</p><h3>Previous editions:</h3><ul><li><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-0-5b630aa34187">The Clevergy´s journey. A startup logbook. Month 0</a></li><li><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-1-74c0b589fe8">The Clevergy´s journey. A startup logbook. Month 1</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d2da0d3fa2b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-2-d2da0d3fa2b">The Clevergy´s journey. A startup logbook. Month 2</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How a product pitch can help your team align on product vision]]></title>
            <link>https://medium.com/clevergy/how-a-product-pitch-can-help-your-team-align-on-product-vision-936e30de8a9f?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/936e30de8a9f</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[product]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[culture]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Alvaro Perez Bello]]></dc:creator>
            <pubDate>Tue, 21 Jun 2022 07:46:23 GMT</pubDate>
            <atom:updated>2022-06-21T07:46:23.375Z</atom:updated>
            <content:encoded><![CDATA[<p>At <a href="https://clever.gy/?utm_source=medium&amp;utm_campaign=blog">clever.gy</a> we have been onboarding new people every week for the past month. We believe that having a <strong>strong product culture</strong> at the company is key to our success. To accomplish this, we empower every employee at the company to become a potential <strong>clever.gy ambassador</strong> with family or friends. At the end of the day, you never know when you are going to meet your next customer, partner, or investor. How do we accomplish that? <strong>By practicing a product pitch.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*-kLJX-OM2EzxGhr7ByBjqQ.jpeg" /></figure><blockquote>Every new employee at clever.gy pitches the product to the rest of the team at the end of their first month on the company to ensure that they fully understand what problem we are trying to solve, where are we heading and what do we want to build.</blockquote><p>A product pitch is a short presentation explaining <strong>how your product solves</strong> customers’ <strong>problems</strong> and <strong>why</strong> someone should <strong>buy/invest in it</strong>. It’s also knowns as an “elevator pitch” because it should fit in an elevator ride. A product pitch has to be <strong>aspirational and to the point</strong>. It needs to catch the audience’s attention and have a clearly defined <strong>call to action</strong>. If it is properly conducted, anyone can become a potential investor or your next customer.</p><p>A good pitch can be structured in four different parts:</p><ul><li>Problem</li><li>Vision</li><li>Solution</li><li>Call to action</li></ul><h3>Focus on the problem</h3><blockquote>⛔️ No one wants to hear about your new app</blockquote><p>There are more than <strong>2.11 million</strong> apps in the App Store and a lot of buzz about the next Spotify or Netflix. If you want to draw the audience’s attention don’t start with your app. Going over the features, the screens, and the awesome algorithms that you are developing are not going to get you there.</p><p>If you want people to become invested in your product, first you need their buy-in on the problem that you are trying to solve.</p><blockquote>You need to clearly state the problem that you are trying to solve.</blockquote><p><strong>Providing numbers</strong>, telling <strong>personal stories</strong>, or even <strong>joking</strong> about the pains of the problem are great ways to <strong>engage your audience</strong> <strong>with the product</strong>. Everyone has problems but if you state them in simple terms the audience will be more likely to become engaged for the rest of the presentation.</p><p>As a company, we need to have a clear vision of the problem that we are trying to solve. Marketing, sales, product design, engineering, HR, … The whole team needs to understand why we are coming to the office every day. There is a great video from Ash Maurya about how great companies focus on problems instead of solutions.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FfS46W0xKrGA%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DfS46W0xKrGA&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FfS46W0xKrGA%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/948a8f5bc1ea755811ddfa57cc829959/href">https://medium.com/media/948a8f5bc1ea755811ddfa57cc829959/href</a></iframe><blockquote>Defining and communicating a painful problem helps us connect with the audience</blockquote><h3>State the vision clearly</h3><p>You’ve done it, you can see it on their faces. They are raging! Oh, that problem! Curse that problem! Why did you even mention it? You are now on the same page, the problem is clear and they want it to be solved. It’s time to pull out the big guns. <strong>Sell them a world where the problem is gone</strong>. All the pain is substituted by an amazing experience. A world so incredible that could fulfill their wildest dreams.</p><blockquote>🌈 The vision shouldn’t contain the company or the product on it. It’s a promise of a better world.</blockquote><p>There is a lot of controversy regarding a company’s vision and mission. When stating the vision in the pitch forget about the company’s wording and focus on the bigger cause. When I speak about this topic, I like to refer to this video of Simon Sinek where he defines the concept of <em>Just cause.</em> Helping your potential customer relate to this <em>just cause</em> will increase their level of empathy for the product.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FzpzZumZCdWA%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DzpzZumZCdWA&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FzpzZumZCdWA%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/f315ec9b68722ec256859cac84cf59e7/href">https://medium.com/media/f315ec9b68722ec256859cac84cf59e7/href</a></iframe><blockquote>😍 Telling an inspiring vision will help you sell your product in the next phase of the pitch</blockquote><h3>Explain how the vision is full-filled with the product</h3><p>This is the tricky part, the customer has been through a rollercoaster of emotions, and you want to leave them wanting more. Remember, no one wants to hear about your new app unless it gets rid of that nasty problem. It’s important to be pragmatic in this part. <strong>Don’t spend a lot of time on the features and services</strong> that your product provides, just <strong>show how your product leads to the vision</strong> previously sold. Unless the audience is techie, they won’t care about your microservices architecture, your front-end stack, or the blockchain you are running your product on. All they need to know is what it does and how to get it.</p><blockquote>📢 Structuring the functionality of the product in the same way that you structure the problem, will help your audience understand your product benefits faster</blockquote><h3>Include one call to action</h3><p>The elevator stops and the person smiles back — Thank you for the story — leaning towards the exit of the elevator. Your one last chance.</p><p>If they:</p><ul><li>Don’t relate to the pain points you described.</li><li>Don’t care for the vision you stated.</li><li>Don’t think the product is a good idea.</li></ul><p>It’s ok, you gave it a shot, took your chances and you missed. It is not a big deal!</p><p>But if at the end of the ride they:</p><ul><li>Engaged with the problem.</li><li>Were inspired by the vision.</li><li>And wanted to buy the product.</li></ul><p>It would be horrible if you let them go without the opportunity to invest in it.</p><blockquote>‼️ That’s why the call to action is the most important part of the pitch</blockquote><p>As an ambassador of the company, what goal do you want to achieve when the elevator reaches the top? Is it his contact information to send a follow-up email? Do you want feedback? Do you want a new user to download the app?</p><p>Think about what you want to achieve with the conversation.</p><blockquote>💡 A well-defined call to action will turn an informal conversation into meaningful opportunities</blockquote><h3>Now let’s pitch why employees should pitch the product pitch to the team</h3><p>Running out of cash is the number one problem that startups face today. It is often the case that money is allocated to noncritical activities. Day-to-day decisions can make the company drift away from priorities and over time lead to bankruptcy.</p><p>What if the whole team could look out for the company’s best interests? Cutting expenses shouldn’t be the answer, but to improve the product and sell it. Imagine the marketing team generating fewer but more qualified leads, product design delivering fewer items for developers to develop, and developers solving mainstream problems instead of edge cases. And most importantly, picture your whole team validating ideas and selling through family and friends. That would be the dream.</p><p>By ensuring that every member of the company has a good grip on what problems the product tackles and the company’s goals, their daily decisions will be more aligned across teams. Although preparing a good product pitch takes valuable time from the first days of work, it pays off in a long-term vision.</p><p>So what are you waiting for? Start making your team pitch your product to ensure you meet your next challenge!</p><p>You can read more about how we create a healthy product culture at <a href="https://clever.gy/?utm_source=medium&amp;utm_campaign=blog">clever.gy</a> in our Monthly log book.</p><p><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-1-74c0b589fe8">The Clevergy´s journey. A startup logbook. Month 1</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=936e30de8a9f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/how-a-product-pitch-can-help-your-team-align-on-product-vision-936e30de8a9f">How a product pitch can help your team align on product vision</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Clevergy´s journey. A startup logbook. Month 1]]></title>
            <link>https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-1-74c0b589fe8?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/74c0b589fe8</guid>
            <category><![CDATA[cleverly]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[strategy]]></category>
            <dc:creator><![CDATA[Juan López López]]></dc:creator>
            <pubDate>Tue, 31 May 2022 05:31:37 GMT</pubDate>
            <atom:updated>2022-05-31T05:31:36.968Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*PIIfLETIbEb8-_du" /><figcaption>Photo by <a href="https://unsplash.com/@e4xtream">@e4xtream</a>, unsplash.com</figcaption></figure><h4>For how many centuries have we been working at Clevergy?</h4><p>OMG, It feels like we’ve been working at Clevergy all our lives. When a company is starting a lot of things happen and that intensity can be deceptive. <strong>Weeks can seem like days</strong>.</p><p>But going back to the real world. Imagine that you are starting a startup, it is your second month, the ideas are endless and the work even more so. And worst of all, money is not. <strong>You can go crazy, or you can start prioritizing</strong>.</p><p><strong>And what is the only way to prioritize? To have a </strong><a href="https://www.goodreads.com/book/show/11721966-good-strategy-bad-strategy"><strong>strategy</strong></a><strong> with clear goals.</strong></p><p>Part of our strategy is building a modular product and we want to do that by working with design, product people, and engineers. These two keys have guided us to prioritize our first tasks. The other guiding light for our strategy is to align with our customers, so product iterations focused on improvement are a constant in our daily work.</p><p>So here you have the different things we were working on after <a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-0-5b630aa34187">month 0</a>:</p><ul><li><strong>The magic trident</strong>. We strongly think that for building a product like <a href="http://clever.gy/">Clevergy</a>, a product role, a product designer and engineers need to work closely. That’s why we have hired a <strong>new product designer and a new software engineer </strong>(she will start next month). In case you didn’t know, we are still hiring engineers, so please contact <a href="/service/mailto:juan.lopez@clever.gy">me</a> if you are interested! (4-day weeks and an amazing team).</li></ul><figure><img alt="Atlassian triad model" src="https://cdn-images-1.medium.com/max/1024/0*WLqD-c0eut2OdAW-.png" /><figcaption>Atlassian triad model</figcaption></figure><ul><li><strong>First AI model running</strong>. It’s amazing how technology evolves. Since I wrote about <a href="https://medium.com/@juaneto/continuous-deployment-in-machine-learning-systems-a799d661dcc3">continuous deployment in machine learning</a> three years ago, cloud providers have worked a lot in the <a href="https://ml-ops.org/">MLops</a> field. After some data wrangling (the ugliest and heaviest part of data work that nobody wants to talk about) and training some models in our computers, <strong>with some clicks we have been able to (in the cloud) train (with big amounts of data) and deploy a new model in little time</strong>. In our case, we use <a href="https://cloud.google.com/vertex-ai">Vertex AI</a> in GCP: an incredible experience. <strong>Thanks to it, you can work in an agile way in AI tasks</strong>.</li><li><strong>We improved our </strong><a href="https://app.clever.gy/"><strong>prototype</strong></a>. We have a <a href="https://app.clever.gy/">public prototype</a> in order to get feedback from people. We don’t want to develop another energy dashboard that nobody understands, <strong>we want to do something that helps people save energy</strong>.</li></ul><figure><img alt="Public demo screenshot" src="https://cdn-images-1.medium.com/max/772/1*isZMbUTFRvzaVWl2gCGiuQ.png" /><figcaption><a href="http://app.clever.gy/">Public demo</a> screenshot.</figcaption></figure><ul><li>We started building <a href="https://micro-frontends.org/"><strong>microfrontends</strong></a>. A part of our strategy is to have a modular product easy to use. There is nothing better than microfrontends to achieve that. Whenever we work more on that, we will tell you about our experience.</li><li><strong>Remote first culture involves async communication</strong>. We want to do this in the best way from the very beginning. We use <a href="https://www.notion.so/"><strong>Notion</strong></a> to document our processes, product, and tech. Regarding this last part, we use <a href="https://adr.github.io/"><strong>ADRs</strong></a> to document our architecture decisions and <a href="https://diagrams.mingrammer.com/"><strong>diagrams as code</strong></a> to document our architecture. We’re also giving <a href="https://www.claap.io/product"><strong>Claap</strong></a> a shot; we believe that short videos with interactive feedback will help us become more involved in decisions.</li><li>We started in a crazy way, and, we realized that we need to work on <strong>organizing</strong> ourselves. We have defined simple <strong>ways of working</strong> that we hope will help us in our daily work.</li></ul><p>Stay tuned for the next edition!</p><h4>Previous editions:</h4><ul><li><strong>Month 0</strong>: <a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-0-5b630aa34187">https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-0-5b630aa34187</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=74c0b589fe8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-1-74c0b589fe8">The Clevergy´s journey. A startup logbook. Month 1</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Clevergy´s journey. A startup logbook. Month 0]]></title>
            <link>https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-0-5b630aa34187?source=rss----630073e22042---4</link>
            <guid isPermaLink="false">https://medium.com/p/5b630aa34187</guid>
            <category><![CDATA[strategy]]></category>
            <category><![CDATA[cleverly]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Juan López López]]></dc:creator>
            <pubDate>Tue, 03 May 2022 05:45:01 GMT</pubDate>
            <atom:updated>2022-05-31T05:55:50.329Z</atom:updated>
            <content:encoded><![CDATA[<h3>The Clevergy´s Journey. A startup logbook. Month 0</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*2A1dnnwlysKvJlXt" /><figcaption>Photo by <a href="https://unsplash.com/@tateisimikito">@tateisimikito</a>, unsplash.com</figcaption></figure><h3>Hello, world!</h3><p>Welcome to the first chapter of our new blog. It is our <strong>first month</strong>, so we have not fully defined our <a href="https://37signals.com/">values</a> (we hope to do that soon). But if we have one thing clear in <a href="https://clever.gy/">Clevergy</a> it’s that we are not afraid to share what we do and how we do it. We think that explaining our ideas, processes and how we do things will never be bad for us: in most cases, it will inspire people to help us. <strong>The community responds to you when you are open with them</strong><em>.</em> I guess it’s that magical word that we call <strong>transparency</strong>.</p><p>Our aim with this logbook is to tell everybody about our journey. <strong>The journey of a startup from scratch</strong>. Specifically, we want to focus on the tech world, since it is interesting to talk about the thousands of trade-offs that you need to consider to successfully advance.</p><p>Something you learn after years in the tech world is that the most important thing when you are starting is not the idea, nor is it the deep knowledge of the topic of your startup:</p><blockquote>The most important thing is the team you have and the skills they bring to the table. While the strategy you create with this asset is important, the ability to responsively and actively iterate solutions and new ideas is invaluable because you will be growing and adapting hundreds of times.</blockquote><h3>Month 0</h3><p>Month 0 in a startup is like when a writer has <a href="https://writingcooperative.com/the-blank-page-syndrome-6e6e5d1250fe">blank page syndrome</a>. You have hundreds of ideas but you need to start somewhere.</p><p>Ultimately, we’ve spent our time in some pretty great ways:</p><ul><li><strong>Winning a startup contest</strong>. WHAT AN INCREDIBLE TEAM WE HAVE. Last Monday we participated in the first edition of the #ConnectingEnergy event, Naturgy’s startup incubator, and our project was chosen as the winner! 🎊. As a result, we are going to have the opportunity to pilot our tool for Naturgy.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RvTi6vMgkb8k4skMEmVTkA.png" /><figcaption>#ConnectingEnergy Naturgy´s event</figcaption></figure><ul><li><strong>Building a strategy</strong>. For us, having a strategy behind the mission of the company is crucial. We will talk about this strategy deeply in another post in the future.</li></ul><blockquote>“Good strategy almost always looks this simple and obvious and does not take a thick deck of PowerPoint slides to explain. […] . A strategy is like a lever that magnifies force. Some fundamental sources of power used in good strategies: leverage, proximate objectives, chain-link systems, design, focus, growth, advantage, dynamics, inertia, and entropy.“ <a href="http://goodbadstrategy.com/">Good Strategy/Bad Strategy: The difference and why it matters.</a></blockquote><ul><li><a href="http://app.clever.gy/"><strong>Building our first prototype</strong></a>. How amazing it was! To have a lean mindset is awesome. Just weeks after really beginning of the company, we put in production the features of our first product last Friday. We wanted something to test with real users and to get fast feedback and metrics. <strong>Doing something fast is not the opposite of doing something well. For us, the important thing is to create an </strong><a href="https://evolutionaryarchitecture.com/"><strong>evolutionary architecture</strong></a><strong> and not lose focus on the fact that </strong><a href="https://martinfowler.com/articles/is-quality-worth-cost.html"><strong>quality</strong></a><strong> is non-negotiable.</strong></li></ul><pre>How do we create this prototype? We have chosen a sample dataset from a user persona we think is the best to experiment with, and built a dockerized React app and put in a managed serverless GCP service (<a href="https://cloud.google.com/run">Cloud Run</a>). Regarding technologies, React and Kotlin debuted as languages in Clevergy.</pre><ul><li><strong>Hiring</strong>. Is there something more difficult in the world than hiring in tech? Unlikely. We are fortunate that our people and our culture help a lot. <a href="https://www.forbes.com/sites/forbescoachescouncil/2020/04/14/hire-slow-and-fire-fast-why-leaders-should-heed-this-advice/">Hire slow-fire fast</a>, because we want to be sure we are creating the team and culture we want. <strong>BTW, if you want to join this beautiful project, ping </strong><a href="/service/mailto:juan.lopez@clever.gy"><strong>us</strong></a><strong>, we are hiring!</strong></li><li>Talk with <strong>universities</strong>. When you are building over the state of the art, nothing better than talking with people who know more than you. In our case, we are talking about possible collaborations with universities because we think there are several win-wins we can achieve together.</li></ul><p>Stay tuned for the next edition!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5b630aa34187" width="1" height="1" alt=""><hr><p><a href="https://medium.com/clevergy/the-clevergy-s-journey-a-startup-logbook-month-0-5b630aa34187">The Clevergy´s journey. A startup logbook. Month 0</a> was originally published in <a href="https://medium.com/clevergy">Clevergy</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>