How can you work with stakeholders to define data quality for process automation?

Según mi experiencia comprender el proceso y sus objetivos es la principal herramienta para continuar con la automatización del proceso y garantizar calidad de los datos. - Definiendo entradas y salidas con sus pasos. - Definiendo el camino feliz y resultados esperados. - Teniendo en cuenta las excepciones que puedan llegar a producirse y su solvencia. - Evitando redundancias en los datos para que no se dupliquen variables. - Dise?ando caminos mas directos para conseguir el output final. En definitiva comprender el proceso implica analizar todos los puntos necesarios para una automatización eficiente del mismo y sus datos.

Cito aqui alguns dos procedimentos para manter a qualidade dos dados nos projetos: - Tenha clareza dos objetivos - Tenha clareza de que você e o stakeholder est?o na mesma página - Fa?a perguntas - Crie métodos para testes - Crie métricas de avalia??o - Fa?a use-case mesmo no desenvolvimento - Melhore sempre que possível

One thing I found effective is to involve stakeholders early in the process. By including them in discussions about data quality and requirements. Actually, I disagree with assuming a one-size-fits-all approach to data quality. Different stakeholders may have varying priorities and definitions of what constitutes good data quality. It's crucial to tailor the definition to each stakeholder group's needs. As an example, I have seen organizations conduct workshops or interviews with stakeholders to gather their perspectives on data quality. This collaborative approach fosters buy-in and helps create a shared understanding of data quality requirements for process automation.

Nesse tempo a frente da manuten??o e automa??o, que já somam 18 anos, aprendi que antes de tomar qualquer decis?o, seja ela para melhoria ou contestar qualquer processo, o primeiro passo é entender bem de perto o funcionamento do processo, sabendo a condi??o atual e a condi??o perfeita. Mais porque isso? Em processos que envolvem variáveis químicas por exemplo, uma altera??o momentanea da matéria prima, pode acarretar a altera??es que n?o surtem efeito por longo prazo, pois em curto período de tempo podemos ter que nos movimentar novamente devido a estabiliza??o da matéria prima. No processo de a?úcar e etanol, por exemplo temos cenários de altera??o a curto prazo da matéria prima. O segredo é ter previsibilidade e antecipar os ruídos.

Na perspectiva de uma empresa de TI, é crucial enfatizar a colabora??o estreita com as partes interessadas ao definir a qualidade dos dados para automa??o. Entender profundamente as necessidades do cliente (interno ou externo), adaptar os padr?es de qualidade de dados às especificidades do área impactada/processo impactado e manter uma comunica??o transparente s?o elementos essenciais para o sucesso na automa??o de processos. Isso n?o apenas fortalece a confian?a com os clientes, mas também impulsiona a eficácia e a relevancia no dinamico cenário de tecnologia.

1. Identify and engage with key stakeholders who have a vested interest in the data and its quality. This may include business analysts, data scientists, process owners, IT professionals, and end-users. 2. Work closely with business stakeholders to understand the overall business objectives and how data quality contributes to the success of process automation initiatives.

Explain the value, the purpose to all the parts that need to be involved. What's in for them in the Data Automation and why the right quality of data is important, to end users, process owners, main stakeholders. Test and show value proposition, minimum product value as soon as possible, to keep momentum and engagement.

En lo personal, considero que el éxito de la automatización de un proceso, radica en la rigurosidad con la que se plantea, la exigencia con la que se prueba y la rapidez con la que se implementa. Para esto es vital tener un feedback constante de los usuarios due?os de la solución, mediante el levantamiento de datos de entrada con calidad, la ejecución de pruebas funcionales y no funcionales, y las salidas a producción planificadas con sus respectivos planes de contingencia para su corrección y ajuste. Laboralmente he podido ratificar que los desafíos técnicos de las automatizaciones, se han resuelto conformando anticipadamente equipos cross funcionales, y los beneficios, se han obtenido aplicando metodologías ágiles entre TI y el negocio.

People, processes, and technology, exactly in that order. Attract people to the project, inform everyone about the expected outcome, and involve everyone in the transformation journey.

One of the highlights of process automation is data gathering for business requirements. Typically, at least two stakeholders are invited to finalize the requirements, providing their feedback and expectations. As a Business Analyst, the best way to engage with your stakeholders is to share ideas from previous projects that you can apply to the current one. This approach allows you to identify potential pain points and determine the most effective method for creating an efficient system.

Something i have noticed - Standards have to be well defined and documented - Metrics to measure data quality have to also be well aligned and defined, so that there is no ambguity.

Data quality is pivotal in process automation, impacting the reliability of automated systems. Collaborate with stakeholders to set clear data standards and governance practices. Establish continuous monitoring and feedback mechanisms to address issues promptly. Implement training sessions to enhance awareness and accountability for data quality. Define KPIs for regular evaluation and document protocols for data validation, ensuring seamless integration and fostering a culture of high-quality data in the automation process.

Automatisiert werden kann fast alles, was irgendwie mit Daten zu tun hat. Nur ob es immer Sinn macht, zeigt sich erst, wenn man sich die Eingabe- und Ausgabemenge anguckt. Besonders, wenn die Datenqualit?t schlecht ist, kann die automatische Verarbeitung dafür sorgen, dass sie mindestens nicht - im Gegensatz zur menschlichen Verarbeitung - schlechter wird. Im besten Fall gelingt die Aufwertung durch Anreicherungen weitere Daten. Allgemein kann man sagen: Je valider die Daten sind, desto h?her ist der Nutzen, den die Mitarbeiter oder ein Algorithmus am Ende haben. Demnach ist die Definition von Datenqualit?tsdimensionen und -indikatoren ein wichtiger Bestandteil im gesamten Prozess.

Start with clarifying the scope, objectives and priorities of the automation process with stakeholders. Then verify and ensure the data source and quality. As the data scientists say: your data has to “ROCC”, that is to be Reliable, Original, Credible and Cited. We should not underestimate the cost of poor-quality data in business, as the figure is shocking! According to IBM, the yearly cost of poor-quality data is $3.1 trillion in the US alone.

Desde el primer momento con la automatización se califica el dato con seguridad de que no va a haber equivocación en el mismo, pues la máquina siempre va a insertar el dato que le digamos sin tener en cuenta el error humano.

One consideration is the potential gap between defining data quality thresholds and achieving them in practice. Establishing realistic and achievable thresholds is crucial, but without effective data governance and continuous monitoring, there is a risk of falling short in maintaining the desired data quality levels over time.

Devemos analisar os resultados obtidos e criar uma boa comunica??o entre as partes envolvidas, para avaliarem métricas, indicadores e atuarem nos pontos de melhoria, consolidando os limites e metas previamente estabelecidos

Identifique falhas e pontos de melhoria: Trabalhe com as partes interessadas para identificar problemas nos dados existentes. Isso pode incluir inconsistências, erros ou falta de integridade. A colabora??o é essencial para entender as áreas que precisam de ajustes.

A partir de los primeras pruebas de la automatización terminada se pueden medir los resultados finales comparando la situación antes y despues de la automatización. Los datos hablarán por sí solos, indicando la mejoría del proceso.

Por que é vital estabelecer limites e metas com as partes interessadas? Transparência e Consenso: Estabelecer limites e metas em colabora??o com as partes interessadas promove transparência e cria um consenso sobre o que é considerado qualidade. Adequa??o às Necessidades: O envolvimento das partes interessadas garante que os limites e metas definidos estejam alinhados às necessidades específicas do processo e da organiza??o. Engajamento Contínuo: A defini??o conjunta de limites e metas mantém as partes interessadas engajadas, proporcionando uma base para avalia??es contínuas e aprimoramentos.

Implementing data quality checks and controls is essential for ensuring that data used in the automation process meets the required standards of accuracy, consistency, reliability, efficiency and compliance. By integrating data quality controls checks in the automation workflow (i.e adding data validation rules, implementing error handling mechanisms, and exception reporting in the automation ), inconsistencies, errors, and discrepancies can be identified and addressed in real-time, contributing to the overall effectiveness of the automated processes. Implementing data quality controls is an essential component of ensuring that the automation initiatives produce reliable result that are aligned with business objectives.

Las comprobaciones de calidad de los datos son para supervisar la calidad de los datos respecto a objetivos. Comparando datos automatizados y manuales a lo largo del ciclo de vida de sendos procesos.

Por que é crucial implementar verifica??es e controles com as partes interessadas? Coopera??o Efetiva: Implementar verifica??es e controles em conjunto promove uma coopera??o efetiva, unindo esfor?os para garantir a qualidade dos dados. Identifica??o Proativa de Problemas: A colabora??o facilita a identifica??o proativa de problemas, permitindo ajustes antes que impactem negativamente os processos automatizados. Sustentabilidade: Verifica??es contínuas mantêm a qualidade dos dados ao longo do tempo, proporcionando uma base sustentável para a automa??o de processos.

The fifth step involves implementing a combination of manual and automated data quality checks and controls throughout the data lifecycle. These actions verify and maintain data quality indicators, ensuring alignment with established thresholds and goals. Emphasize a modern approach by integrating advanced automated technologies for efficient and proactive data quality management.

Além de implementar verifica??es e controles, adote sistemas que possam se adaptar dinamicamente às mudan?as nos requisitos de qualidade de dados, procure solu??es que utilizem Inteligência Artificial para aumentar a efiucácia dos controles. A automa??o de controles, combinada com a capacidade de ajuste em tempo real, garante uma resposta ágil às evolu??es no ambiente de dados, mantendo a qualidade ao longo do tempo.

Dados, toda a??o deve estar baseada em indicadores baseados em dados. Quando falamos em automa??o de processos n?o temos espa?o para achismos ou suposi??es. é preciso entender se de fato é uma necessidade para vislumbrar melhorias que sejam efetivas e realmente sejam uma solu??o e n?o apenas tirar o problema de um lugar e colocar em outro.

Escolha solu??es adequadas para a automatiza??o: Com base nas necessidades e nos pontos de melhoria identificados, selecione ferramentas e tecnologias que garantam a qualidade dos dados. Isso pode envolver revis?o regular das métricas de qualidade, ajuste das políticas de governan?a de dados e busca de maneiras de melhorar a qualidade dos dados...

Here it is critical to have a Data Governance mechanism in the organization. This will help ensure that data quality reviews as captured in metrics and dashboards and other reports are given the right importance and instances of bad data are corrected or dealt with appropriately. The composition of the Data Governance councils should involve Product Managers, Business and Technical stakeholders. With the role of good data becoming increasingly important especially in Generative AI, it is important to provide good clean data as an input for decision making and for removing any sort of bias.

Bei der Kommunikation der Ergebnisse ist es wichtig, dass dabei ein Abgleich mit den vorher vereinbarten Zielen erfolgt. Gibt es Abweichungen und wenn ja welche? Es ist wichtig die Erwartungen der einzelnen Stakeholder konkret zu kennen und ihnen klare Antworten auf ihrer Erwartungen zu geben. Wenn dies nicht m?glich ist, dann müssen wenigstens die Gründe dafür geliefert werden. Damit zeigt man den Stakeholder das man sie und ihre Bedürfnisse ernst nimmt.

To my earlier point, feedback and continual improvement is key to the adoption and success of a new software / automation program. This should be understood at the beginning, so that success isn't purely measured on a delivery deadline.

- Ensure to involve the clients during every step of the way. Mistakes do happen, and it would be wise to rectify them early. One has to remember that the process automation is a process more than an activity. - The client's requirements might change during the course. Be ready to accept them, and establishing an open communication channel would avoid unnecessary reworks, and wastage of time. Validate their requirements then and there. - The cycle has to be repeated for continuous improvements. Make sure the client understands the importance of it and get them onboard for the whole process.

To define data quality for process automation with stakeholders, identify key data elements, establish quality metrics and thresholds, conduct assessments, implement governance, involve stakeholders in testing, and foster continuous improvement.

The first step is to make sure that you are talking with each other, not past each other. I have seen many situations where two parties were saying the same things but using different terminology. So, they were not communicating. Once I guided them through the process of taking a step back to basics and agree on definitions effective communication started.

Engage stakeholders in a Data Quality session, where you collectively decide the accuracy, completeness, consistency needed for high-quality data in the automation process. Also, assess and iterate based on feedback for continuous improvement in process automation.

Man sollte immer beachten, dass man nicht für sich selbst automatisiert, sondern sich selbst komplett aus dem Fokus nehmen. Jeder hat zwar seine Erfahrungen und Vorgehensweisen, aber am Ende z?hlt, dass der Prozess im Sinne des Anwenders optimiert ist. Wenn der Kunde vehement eine Banane m?chte, bringt es nichts, ihm einen Apfel hinzuhalten und zu sagen, dass der doch viel besser sei. Schlüssel hierzu ist stetige Kommunikation und auch etwas Geduld. Ein Prozess kann sich im Rahmen der Automatisierung durchaus ?ndern - und selbst danach noch. So ein Projekt endet nicht mit der übergabe, sondern lebt auch vom Feedback der Anwender, welche auf dem Weg dahin immer der prim?re Fokus sein sollten.