In this instructional YouTube video, the presenter argues that data storytelling is the missing skill for most aspiring data analysts, who often focus solely on number‑crunching and coding. She outlines eight common data‑visualization and storytelling mistakes—(1) the Skittles effect (overuse of colors), (2) 3D charts, (3) gridline jail (excessive borders and gridlines), (4) illegible legends, (5) truncated axes, (6) Frankenstein dashboards (unstructured collections of charts), (7) passive titles, and (8) lone‑wolf numbers lacking context. For each mistake she explains why it harms comprehension and credibility, cites research or industry standards, and provides a concrete fix (e.g., use a single highlight color, keep charts flat, remove unnecessary ink, use direct labeling, start bar charts at zero, apply visual hierarchy, write action‑oriented titles, and always add comparative context). She then presents a unifying principle: effective analysts must persuade, turning dashboards and slide decks into argument tools that drive decisions. The video concludes with a call to action, promoting mentorship, a free training roadmap, and a teaser for a follow‑up video on building portfolio projects that demonstrate strategic problem‑solving rather than generic dashboards.
Chapters
If you don't understand storytelling, you do not understand data analytics. Most aspiring analysts think the job's all about crunching numbers and just writing code. That's only a fraction of the equation. The other part is communicating what those numbers even mean in the first place and how to make people take action. When I first started as a data analyst, I made every single visualization mistake in the book. Rainbow bar charts, cluttered dashboards. I thought it all looked impressive until it wasn't. But the problem wasn't my analysis, it was my delivery. And in this video, I'm gonna break down the 8 data storytelling mistakes that's killing your credibility and how to fix each one. Because companies aren't paying 6 figures for people to make charts. They're paying for people to translate data into action. And if you wanna master these skills with hands-on mentorship and real projects, check the link in the description to see how I can help you avoid these mistakes and land your first job in data analytics. And stick around until the end and I'll share the one principle that separates analysts who get ignored from analysts who get promoted. Now the first mistake, the Skittles effect. This is when you use a different color for every bar in your chart and it doesn't have any relevance. Red, blue, green, yellow. It looks like a candy exploded on your dashboard. In analytics, color is a tool for emphasis, not decoration. When you highlight everything, you highlight nothing. Studies on cognitive load and data visualization show that adding unnecessary visual complexity leads to slower interpretation, more errors, and even worse decision-making. Every color that you add that doesn't serve a purpose in increasing what researchers call extraneous load, that's mental effort wasted on decoration instead of insight. So the fix, go gray. Make your base data gray and use one bold color to highlight what actually matters. If October sales tanked, make October red, then everything else gray. Now the viewer's eyes goes exactly where it needs to go. You're reducing cognitive load and directing attention simultaneously. This is something that I drill to my students before they even touch Tableau or Power BI. Data storytelling principles transfer across every tool, and if you can't even tell a story in Excel, you won't magically figure it out in a fancier software. Storytelling actually takes longer to grasp than most technical skills. It requires feedback and constant exposure to new information. You can't just watch one tutorial and just get it in a second. A 2024 survey found that mastering data storytelling led to improved stakeholder communication and also better decision-making. This skill moves the needle. Most people underestimate how valuable it is but overestimate their ability to do it. Now, mistake number 2, using 3D charts. Whether it's a 3D pie chart or a bar chart, this is an immediate red flag. Why? Research confirms that 3D effects distort perspective and reduce comprehension. Slices at the front of a pie chart will actually look bigger than slices in the back, even when the numbers are identical. You're basically lying with data without realizing it. All of this actually impairs accurate reading of values. Studies consistently recommend using 2-dimensional charts for statistical graphics and reserving 3-dimensional charts for geographical terrains or molecular structures. The fix is very simple. Just keep it flat. Always. There's really no business scenario where a 3D chart communication is usually better than a 2D one. It's just visual noise that undermines your credibility. I reviewed many portfolios and anytime I see a 3D chart, I know this person probably hadn't studied what professional work looks like. Your portfolio is your proof, and saying that you took a bootcamp is not evidence. Everyone has that these days. Your portfolio separates you from thousands of applicants who claim that they know Tableau and that they know visualizations. So if your project still has 3D pie charts, you're signaling that you still don't understand the basic visualization principles. Okay, mistake number 3, the gridline jail. When you create a chart in Excel or Tableau, it comes with heavy borders, background colors, and dark gridlines by default. Your data looks like it's actually trapped in a cage. I remember my first dashboard. It had every single default setting turned on. Borders everywhere, gray backgrounds, dark gridlines cutting through everything. I thought it looked complete, but it actually looked like a mess. Every element on a chart should have a purpose. If the ink isn't helping someone understand an insight, delete it. Research on dashboard design backs this up. Reducing extra visual elements improves comprehension and decision quality. So remove the borders, remove the background colors, and lighten or remove gridlines. White space is your friend, and so let the data breathe. I always make sure I keep 2 books on my desk, and I always recommend these to my students. Storytelling with Data and Information Dashboard Design. The second one actually covers 13 common dashboard mistakes. I had an actual physical copy on my desk when I had my first data analyst job. Whenever I was ready to build a chart or a dashboard, I always referenced it. There are so many errors we make without even realizing it, and without feedback or self-education, we're just running blind. Okay, mistake number 4, failing the legibility test. This happens when you use legends that force users to have to play eye tennis. They look at the blue slice and then scan to the legend and then back to the chart to match it up. I watch people who do this with presentations and dashboards all the time. Their eyes have to bounce back and forth trying to decode what they're actually seeing instead of absorbing the insight, and that's a design failure. So don't make stakeholders work that hard. They're busy. They have 10 other things competing for their attention. So the fix here is direct labeling. Put the category name or the percentages right next to the data point. No legend required and information delivered instantly. This separates someone who knows tools from someone who thinks like a business leader. Anyone can drag and drop in Tableau, but few people design experiences that guide decisions. If you're watching this and thinking, I have no idea how to actually break into this field, I got you. I created a free training that walks you through the exact roadmap that my students and community of over 80,000 use to land data roles in 6 months or less without going back to school and wasting money on bootcamps and courses that just don't work. Click the link in the description to access that training. If you're ever ready to work with me directly, we can talk about how to fast-track your transition with real mentorship, real projects, and Real results. Okay, now mistake number five: the truncated axis. This is when you start your bar charts at an axis at maybe fifty instead of zero to make small changes look massive. I've seen analysts do this to make their charts look better than it actually was in reality. It works until someone catches you, and they will catch you. If a senior stakeholder realizes that you manipulated the scale, they're not going to really trust you with your analysis again. In data, trust is everything, and once it's gone, it's gone. You become the analyst who. Exaggerates and that reputation sticks. This is documented research on misleading visualizations that happen all the time with politics. Manipulated scales are a common source of misinformation. So it's not just bad practice, it's very deceptive. Bar charts should always start at 0, period. Now there is one exception. Line charts tracking small fluctuations like stock prices or interest rates, you can truncate to show volatility, but label it clearly so no one feels tricked. But for bar charts, always 0. Mistake number 6, the Frankenstein dashboard. I see this constantly in portfolios for aspiring analysts. It's just a canvas of random charts, and some of them are not even good charts, and they call it a dashboard. It's not a dashboard, it's just noise. When I review portfolios and I see this, I already know that the person just isn't ready for an interview or to get hired. They haven't studied what professional work actually looks like, and worst of all, they haven't even gotten real feedback, so they don't even know. They're just throwing things together and hoping that it works. Use visual hierarchy, and research on dashboard design is clear. The best dashboards use progressive disclosure. You want to show your summary stats or KPIs at the top. You want to utilize drill downs for depth, and this all reduces cognitive load while giving users the access that they actually need. All right, mistake number 7, passive titles. Stop titling charts sales by month or customer churn rate. That just describes data, but it does not tell a story. It just tells me what I'm looking at, but not what to take away. You wanna use action titles, so change sales by month to sales peaked in November due to a Black Friday sale. Now stakeholders know what they're seeing and why it matters, and they no longer have to guess. Research shows people remember only about 5 to 10% of information presented as statistics alone, but when that same information is embedded in a story, retention jumps to 60 or 70%. This is also critical for slide decks, and here's something that no one really talks about. Data analysts don't just make dashboards, we make PowerPoint presentations, a lot of them, and the best decks to learn from are consulting-style decks from the Big 4 like BCG or McKinsey. This also matters in interviews. Take-home assessments are very common for higher-paying roles in data. They give you a dataset and a business problem, and you have to do your analysis and create a slide deck presentation. If your presentation looks like junk, you're not going to be invited back for an interview. So create a good deck, present it well so you can get the offer. Okay, now the final mistake, the lone wolf number. This is when your dashboard just shows $5 million in revenue with no context. Is that good? Is it bad? Did we do $10 million last year? Did we expect $3 million? A number without context means nothing. It's just floating there. I see this all the time. Big bold numbers on dashboards with zero reference The person who builds it may know what it means, but everyone else is confused. You always want to compare metrics to something. You always want to have a benchmark— previous year, target goal, industry average. So $5 million revenue, but down 10% versus last year. Now that is a story. $5 million in revenue alone doesn't say anything. And neuroscience actually backs this up. The brain is a story processor, not a logic processor. People hear statistics but they feel stories. When you add context and comparison, you're transforming a number into something the brain can actually process and remember. This connects to thinking like an analyst versus just a technician. A technician just pulls numbers and displays them. An analyst asks, so what? And what decision should someone make based off of this? You're a problem solver first and technician second. You're not paid to just crunch numbers. You are paid to create clarity. All right. So now here's a principle that I promise. The one thing separating analysts who get ignored from analysts who get promoted and hired. Real analysts don't just analyze, they persuade. Your dashboard isn't a report, it's a persuasion tool. Your slide deck isn't a summary, it's an argument. Every visual that you create should be designed to influence a decision. The people who rise fast in data analytics understand this well. They guide what should happen next. They walk into meetings and say, based on my analysis, we should shift the budget from Facebook to LinkedIn because our B2B conversions are 3 times higher there, and here's a 90-day plan. That's influence, and that is persuasion, and that is what gets you promoted. Your job isn't to inform, it's to convince. And when you can master that shift, you stop being just a report builder and you become a strategic partner. A strategic partner gets paid. Okay, now that you understand how to communicate analysis effectively, there's another piece that most people get wrong. You can have the best data storytelling skills, but if you're not building the right portfolio projects, none of it matters. In my next video, I break down exactly how to create projects that get you hired. Not the generic dashboards that everyone else builds, but strategic similar to what I teach my students that proves that you know how to solve actual business problems. Click here to watch next.
