You’re struggling to meet deadlines for data visualization. What’s causing the delay?

The delays may be due to one or more of the reasons such as (in no particular order) - incorrect/ invalid/ missing data (poor data quality) - complexity in relationships between data which has not been properly modelled - not having thought of proper visualisation to represent the insight generated , leading to changes once data is starting to get visualised - not having clarity on what’s the aim of visualisation Before any data is visualised, a clarity is needed on what it is supposed to achieve. Data visualisation should be enmeshed with storytelling to make the insight consumable and achieve the outcome desired. Of course, this has to be backed with a solid data foundation (quality data) and a tool that can deliver the outcome.

Data Quality is indeed something crucial. I often used data visualisation tool to check first the data quality. Checking for outliers or misspelling in categories names is something I always do first for instance. There is no point to build a dashboard with fake data or if 80% of the data is missing because axis of the analysis could be completely different with the complete dataset.

Did you commit to doing a dataviz project by a certain date WITHOUT first getting access to the data? You're in for trouble! Missing data at the start of a project is a huge red flag for me. I would even go so far as to say, don't commit to do a dataviz project unless you've seen the data. Countless projects have been delayed or derailed due to delayed access to real data. Fake data doesn't cut it!

Bad data quality can quickly kill a DataViz project. Having data quality issues is like having a bad foundation for your new house. It's almost impossible to build your house if the foundation is bad. Same for Data Visualization dashboards, load up bad data and your visualizations won't make sense; they won't tell a valid story; and certainly, will not be accurate. All of these will eliminate user confidence, which results in a totally unused dashboard gathering dust in the corner of your Business Intelligence project. Yes, you can use some data cleansing features within the DataViz tool, but I suggest pushing all this data quality cleanup back to the database. That way, clean data will be available to everyone using that data source.

Eu sempre gasto muito tempo qualificando e entendo os dados antes de construir o Dashboard. Quando você faz todo o tratamento e entende bem o conteúdo, o tempo que você gastou, você reduz em dobro na construção do Dashboard. E nunca trate o dado direto no Painel pois a manutenção fica bem complexa.

Define Clear Goals: Clearly outline your objectives and questions before diving into data visualization. This helps in filtering out irrelevant data and focusing only on what's essential. Use Data Sampling: When dealing with large datasets, consider using data sampling techniques to work with a representative subset of data. This reduces processing time and simplifies analysis.

The complexity of data means you should take time to redefine your data, ask questions about the source to ensure data validity , once you can simplify your data. To your own understanding, then visualization should be very easy.

It seems the complexity of data can be one of the reasons for slowing down the progress with visualizations. Spending extra time, carefully analyzing and designing visuals to make sure they're accurate and easy to understand of course results in delay.

Alongside the factors mentioned, another reason for delays in data visualization could be the challenge of combining data from different sources. This can be tough because each data source might have its own format and quality. Trying to bring them all together can take a lot of time and might lead to mistakes, causing delays. To tackle this, using strong data integration methods, like making sure all data follows the same format and using automation tools, can help speed things up. Also, having good teamwork between data engineers and visualization experts can make the process smoother and prevent delays.

La complejidad de los datos siempre va a existir en una organización madura de data, ya que se va a requerir visualizar información externa a la empresa de distintos lugares de la red para compararse con la competencia y entender a los clientes y/o usuarios, con el objetivo de encontrar oportunidades de negocios.

One debatable question over here - many might want to go grand with their visualization vs some who might want to create simple effective charts that everyone can intuitively understand. It's important to prioritize the latter first in most cases as executive and stakeholders are most interested in getting the key insights to drive their decision making. However, if time permits or the insight warrant, one can explore visuals that enhance the quality of the insight while not overcomplicating it.

Sobre alfabetização dos dados eu sempre busco entender o que a área de negócio já tem de informação pronta, como ela reporta os dados para os executivos e o que falta de dado. Isso faz abrir a mente para entender melhor o conteúdo e poder comunicar de forma mais clara nas visualizações construídas.

Data literacy is essential for creating impactful visualizations that convey your message effectively. However, not everyone possesses this skill. You might face barriers like lack of knowledge, skills, or confidence, or resistance from stakeholders. Improve by seeking training, experimenting with data, and collaborating with peers.

The delay in meeting deadlines for data visualization can be attributed to a few key challenges: The data often arrives in an unclean state, necessitating significant preprocessing and cleaning efforts. There is a recurring issue where stakeholders find certain types of charts and visualizations difficult to understand, which leads to additional time spent on revisions and explanations. The requirements from stakeholders are prone to frequent changes, as they continually seek more information or different presentations of the data, causing further delays.

This is a very important one. It's extremely crucial that the story one tells from their data is the truth, the whole truth and nothing but the truth. It's very easy to tell a misleading story from the data (whether intentionally or otherwise) because of something you want/don't want to see. Data can take on a very different shape when viewed through different lenses, so it's key that you ensure the data gets scrutinised from all angles to give an overall picture that tells an accurate story, and that story is portrayed the correct way

Data Ethics considerations can sometimes cause delays in the data visualization process, simply because this is a stage that would not be wise to rush through. Data governance and other processes around security and access must be followed carefully, to ensure sensitive and private data is not made available beyond its intended audience. We can make this stage more efficient by bearing it in mind from the start of a project, asking questions about where and how a data visualization resource will be shared, with whom, and how the underlying data will be stored.

Data ethics introduces delays in data visualization projects due to the meticulous attention required to ensure compliance with ethical principles. Incorporating this involves tasks such as anonymizing sensitive information, adhering to privacy regulations, & mitigating biases in data representation. Each of these steps demands needs consideration & execution, consequently extending time needed for data preparation & visualization. Despite these challenges, prioritizing data ethics is essential for building trust with stakeholders & safeguarding against potential legal & reputational risks. Investing in robust frameworks, training programs, & automated tools can help streamline process & minimize delays while upholding ethical standards.

Choosing the right tools affects visualization speed and quality. Using outdated or incompatible tools leads to issues like performance and usability problems. To optimize data tools, assess needs, compare options, and regularly update and maintain them.

Existen diversas herramientas de datos en el mercado. Pero, en mi experiencia no puedo recomendar alguna sin antes entender la necesidad a desarrollar, esto implica el volumen de datos a recolectar y procesar, frecuencia de actualización y cantidad de usuarios que consumirán los informes. De esto 3 principales parametros, dependerá la elección correcta para el costo/beneficio de la inversión a realizar.

Oftentimes we, as people, fall into the trap of familiarity and favouritism. There may be a handful of tools in particular we favour that we feel more comfortable with, but that doesn't mean they're always the best option to do the job. It helps to have access to, and knowledge of, a range of tools to do the same job so you know which ones are better suited to doing the task at hand. Just because something CAN do the job doesn't mean it SHOULD, and that's an important distinction to make when picking and choosing the right tools for the job

When there are frequent changes or revisions to the requirements, this will cause delays. This is especially true when the complexity of visualizations are complex. That's why clear communication and planning ahead is key to any project.

You might face challenges such as getting timely and relevant feedback, managing multiple and conflicting feedback, or implementing and communicating feedback changes. To streamline your data feedback, you should establish a feedback plan and schedule, use feedback tools and methods, prioritize and categorize feedback, and document and report feedback actions.

Data feedback is crucial for improving visuals, aligning them with goals, and measuring impact. Yet, it can cause delays without a clear system. Challenges include timely input, managing conflicting feedback, and implementing changes. To streamline, establish a plan, use tools, prioritize feedback, and document actions.

Data quality is my first thought when it comes to reasons for delays. Clean data is essential for both accuracy and how the visualizations work (or not work). My second thought is to make sure your queries are optimized and the hardware that the tools are utilizing are sufficient to handle the work load.

In one of my previous roles, a senior person in the company insisted that she signed off on all decisions, however minor. This led to delays in delivery and frustration on my part as she was often unavailable to speak to. In this scenario, it is important to flag the issue early (without apportioning blame) and suggest solutions, such as delegation, asynchronous feedback or extending deadlines.

Before you start: 1. A sprint 0 kick off. With the FULL team. 2. You understand: Roles & responsibilities. Identify key stakeholders. Establish 1 person to filter / decision delegate. 3. You have access to: The client. The users. 4. Make explicit that intuition or guess work is exactly that. You need to test ideas against predefined criteria. 4. You have: The data. The content. A data dictionary / glossary of terms. 5. If working with a Client Service Team / Engagement team / or any other representatives you FULLY understand what they have promised (without your knowledge or consent). 6. The deliverables, required tech, existing tech infrastructure, handover requirements, ongoing maintenance requirements, IP concerns & any / all timings.

Delays in meeting data visualization deadlines might be due to unclear requirements, along with problems in clean data warehouse design and data quality gaps. This leads to additional time spent on data cleaning and preprocessing before visualization can begin. 📊⏳ #DataViz #DeadlineDelays

Frequently, after finalizing a dashboard or tool, businesses often request alterations once they've commenced utilization during the validation phase. This recurrent scenario significantly contributes to individuals' challenges in meeting deadlines for data visualization.

Outro ponto que vejo muito que causa atrasos na Visualização (etapa final), é relacionado ao entendimento da necessidade do cliente e o entendimento do público...

The main solution I see is accurately estimating the time required to develop the necessary data visualizations on the front end. Teams that practice thorough task estimation, such as in a scrum pointing session, tend to fare better in this regard. By consistently evaluating these estimations, we gain valuable insights, such as realizing it might take longer than initially thought to complete a dashboard from scratch (ie "Wow, it doesn't take us two weeks to complete a dashboard from scratch, it is actually a little bit more"). Conversely, we might discover efficiencies (ie"Wow, it only takes the team an hour to add a field to an existing dashboard").