In this 2026 YouTube presentation, Kedeisha Bryan, founder of the Data in Motion community, warns that the data‑analytics job market remains a "gold rush" but the rules have changed. Hiring standards have risen: companies now seek analysts who can define business problems, translate data into actionable recommendations, and communicate insights effectively, not just those who can run SQL or build dashboards. Bryan outlines a step‑by‑step career strategy: first adopt a problem‑solving mindset using the DMAIC (Define‑Measure‑Analyze‑Improve‑Control) framework; then master a core technical stack (Excel, Power Query, SQL, Tableau/Power BI); next develop data‑storytelling skills and consulting‑style slide decks; and finally, build a differentiated portfolio by delivering real‑world consulting projects (e.g., 30‑day pro‑bonos for nonprofits). She cautions against over‑reliance on generative AI, positioning AI as a "copilot" for mechanical tasks while emphasizing that the analyst must retain logical judgment. The talk also includes a Jeff Bezos anecdote illustrating the need to align metrics with anecdotal evidence, and it offers practical advice for interview take‑home assessments. Bryan concludes with a call to action: follow her detailed roadmap (linked in the video) to become part of the top 1 % of data‑analytics candidates.
Chapters
In the next 12 months, one of two things is going to happen to your career. Either you're going to learn how to leverage data analytics and AI to become irreplaceable, or you are going to watch someone less qualified and talented than you take the opportunities that you deserve. Because right now, there is still a gold rush happening in data, but the rules of the gold rush has completely changed. But most people are trying to break in by playing by the old rules. So let me show you exactly what going all in on data analytics and AI actually looks like this year, because it's not what you think. If you're new here, I'm Kedisha. I am the founder of Data in Motion, a community of over 80,000 current and aspiring data professionals. And I've helped thousands of other career changers break into data analytics without going back to school, got CS degrees, and landing roles at Fortune 500 companies. I have watched this field transform in real time. And what I'm seeing right now is one of the biggest shifts that I have seen since I broke into the field from a delivery driver to 6-figure jobs. Okay, so the gold rush is still on, right? That's still real. Companies are still drowning in data, desperate for people who can make sense of it. But here is what most people don't realize. The for actually getting in and getting hired has gone up significantly. Years ago, you can just learn any SQL tool, any BI tool, and build a portfolio, and you can just get hired pretty quickly. That path still works, but it takes way more now. Meta is laying off 8,000 people, not because they're struggling— obviously they're not— but because they're replacing one type of worker with another type. And they're not the only ones. Companies across every industry are restructuring their teams around these higher standards. Okay, so what does this mean for you? It still means that companies are still hiring for analyst roles. It's just that they're not looking for what most people are are thinking that they're looking for. They're looking for that so-what person, the person who can walk into a room, present an analysis, and make a recommendation that actually changes how the business operates. That is the new standard. Everything else I'm about to tell you is built around that. The second major shift is AI, and this is one that's going to surprise you because the danger isn't that AI is going to be taking your job. Everything you hear about that is a lie. The danger is what it's doing to the people trying to get into this field. What I'm seeing right now from aspiring analysts is very alarming. People are using ChatGPT to write their entire queries, building their entire dashboards, and generating their entire analyses and then putting all of that in their portfolio and in their reports. And that's what they think leveraging AI is all about, but it's not. That's outsourcing your brand and your strategic thinking, and that is creating a massive gap between people who can actually think and people who just know how to prompt. Companies are actually catching on to this. I'm going to show you exactly how they're testing for this later on in the video, but just know this: the people who are winning right now are not those who use AI the most. They're the ones who know when stop using it. And here's the part that should actually excite you if you're coming from a non-technical background. This field is designed for career changers, maybe more now than ever, because if companies are hiring for judgment, business context, and the ability to communicate insights to non-technical people, your industry knowledge combined with data skills, that is a combination that companies are struggling to find and AI cannot replace. All right, so let's start getting specific. What exactly do you need to focus on to capitalize on this shift? Because most of the advice that you're hearing online is either outdated, flat out wrong, or bought and paid for by some sponsor. And I'm not just going to tell you, I'm going to show you. All right, so first off, independent problem solving comes first before any tool. Before you touch any software, you need to understand how to break down a problem. What is the actual question? How would you measure the process? What data do you need? And what would a good answer look like? I like to use a framework called DMAIC— Define, Measure, Analyze, Improve, Control. It comes from Six Sigma. It's simple because it forces you to define the problem before you even open a dataset. Most people skip this entirely. They just jump into exploring data with no question in mind. That is not analysis. That is just scrolling through Columns. All right, so let's walk through a DMAIC problem step by step using an example of being a data analyst at a hospital. So let's just say for step one, for our define, we see that the problem is 30-day readmissions jump from 12% to 18%. So why the heck did it jump 4 points? So what we're going to do is first we're going to have to really measure this process of 30-day readmissions, but we need to pick measures all about picking the metrics, right? So what is the readmit rate? Time to readmit. Whenever you're doing averages or medians, statistics, this is where it comes in, your measure. Then you're going to analyze. So after you know the metrics that you're going to use and analyze for, then you're going to create charts. And this is an example. So here, if you create a chart and you can clearly see that the top diagnosis that is contributing to the high readmissions rate. So one thing that we can see with this Pareto chart, there's practically the top 1 through 4, or just a very few amount that's contributing to 80% of the readmissions rate. And so something like that, you take and you go to the improve stage and improve is all about providing the recommendations. So anytime you're doing your courses, your capstones, whatever, and the prompt is, hey, provide recommendations. It is all about how do you take what you found out in the analyze phase in your charts and provide recommendations to always improve process. Remember, we're always just analyzing business processes. Everything can go boil down to a business process. And whenever you provide recommendations, it's all about improving something, whether it's reducing risk or improving revenue, whatever it is. And then control. And so this is where the control step is where your dashboards come in. So again, as analysts, we are analyzing business processes and dashboards visualize business processes. So Again, dashboards are not just pretty charts, but this is where the decision makers can pretty much see whether a process is normal, whether it is out of bounds or abnormal, whether there's outliers, whatever it is. But this gives them the visual view so that they can see this and then they can take this and go to the next steps. Jeff Bezos had a great example for this. During the days of Amazon, their metrics showed that customer service wait times were under 60 seconds, but customers were complaining nonstop that the wait times were too long. His team
So one day we're in a meeting and we get to this metric in the deck. The guy who leads customer service isn't fitting the metric. And I said, okay, let's call. And I dialed the 1-800 number and called customer service. And we just waited in silence more than 10 minutes.
That is when the metrics but the anecdotal data was not lining up. That is what an analyst does that AI cannot— being able to question the metric, not only creating metrics but lining it up with the anecdotal data. And that is what Kedeisha Bryan does. Anecdotal data. Whenever they don't match, investigating why. Okay, next up is a core technical stack that is non-negotiable. So Excel for pivot tables, Power Query, descriptive statistics, SQL for joins, CTEs, window functions, aggregations. This is your technical foundation that is still getting tested in interviews. But here's where most people are falling short: data storytelling. And I don't just mean making charts. I mean actually knowing the difference between a good visualization and a bad one. Let me show you. So let's take a look at the difference between what a great dashboard looks like and what a not so great dashboard looks like. So when we first start learning Tableau and Power BI, all of our work looks like this. Now, the issue is most of us stay here and not actually know what great looks like. And we put these types of dashboards on our portfolios. We move on to SQL and Python. With a skill set such as this, and we don't really improve our storytelling and dashboard building skills. And this is where most of us mess up. But what we ultimately want to do is graduate into good looking, clean dashboards such as this. So a dashboard like this, this is going to be the next level that you should be aspiring to and getting to before you're applying for jobs. If our dashboards are still looking like this. We should not be moving on to another tool until it looks like the previous one. Most people are building dashboards that look like the first one, and that's not storytelling. It's just plain noise. Nobody is making a business decision based off of that. You need to be able to also build consulting style slide decks. This is a completely separate skill from just building dashboards, and it also gets tested through take-home assessments, which are becoming more common, especially for higher paying roles. They'll give you a dataset, a business problem, and a deadline. You build the analysis, the deck, and present it. If your deck looks like a data dump, you're not moving forward. The tool gets you in the room, but the thinking is what gets you hired. All right, so let's get real about AI again, because everyone is telling you to use AI as your pilot, which is completely wrong. They are overstating how effective it's becoming, and they're making it sound like you can just dump a dataset into ChatGPT and allow it to do the job for you. It won't. AI should always be your copilot. You are always the pilot in charge, and that distinction matters more than anything else I'm about to say in the rest of the video. Use AI for the mechanical grunt work: data cleaning, debugging, formatting. Brainstorming, all of that makes you faster and all of that is fair game. But here's where you stop. Do not use AI to write your analysis. Do not use it to generate your recommendation and do not just paste a dataset into ChatGPT and say, ask for insights. And this isn't just my opinion. MIT even published a study that showed AI boosts your output at first, but it weakens your ability to solve problems on your own. The people who relied on it for direct answers got worse over time. And the ones who use it more like a hint system where they still had to do the thinking stayed sharp. So here's the rule. AI does the labor, you do the logic. All right, so this is the last one and what also brings everything together. Every project on your portfolio should start with a business question and end with a recommendation. And a great way to do this and apply everything to a real business problem is to find a local nonprofit and offer 30 days of free analyst consulting. Clean their data, build a dashboard, and deliver a consulting-style deck with recommendations. This becomes part of your portfolio, your case study on your resume, and real experience that you can talk about in interviews. And this is what separates you from everyone else applying for the same role. I have said this in multiple videos and I will continue saying it over and over again because still not enough people are actually doing it. So if you start doing this today, I guarantee you, you're going to shoot up to the 1% of the competition. All right. So I just threw a whole lot out of you. So let's break this back down. The gold rush is still going on, but the bar is just higher. Problem solving before tools, storytelling that drives decision, AI as your copilot, real business projects, not just tutorials. And if you want an exact step-by-step roadmap, what tools to learn in what order, week by week, exactly how to build a portfolio and brand that gets you hired, I made a full video that breaks it all down. The link is right here.
