My Business Metrics Framework (Think Like a Senior Analyst) transcript

In the age of AI, a differentiating factor between analysts who get hired and promoted on the job and those who get stuck in the job market or in their current careers is the ability to tie metrics back to real business impact. And the ability to do that comes down to one main thing, which is understanding what actually drives the business and what metrics you need to track to prove that. So, if you are a working professional or an analyst trying to transition into data and you've been wondering, "How do I know what metrics actually matter for a business and what metrics do I need to show in my resume, my portfolio to land interviews and offers?" Then this video is for you.

We are going to do a crash course on business metrics so you understand how metrics fit into the day-to-day job, how to actually decide which ones matter, and also how to connect these metrics back to real business impact. If you're new here, I'm Christine. I'm a former data director and a hiring manager who now helps working professionals and analysts stand out in the job market and on the actual job. So, how do you know what metrics actually matter?

We are going to use one simple idea to make this a lot less overwhelming. You can think of a company's metrics like a layer cake where there's three different layers, each one holding up the one above it. So, at the very top you have the icing on the cake. This is basically the part that everyone outside the company actually sees and it's often also what creates the fastest impression of how the company is actually doing.

So, in terms of who actually tracks these numbers, that would be the exact anyone in leadership and people on the board, for example, if it's a public company. These are metrics that are focused on the highest level business performance and they are also often shared with the public if it is a public company. Let's go with an example. Pretend you are a product analyst. Then a company-level metric on the product side would be something like active users.

This is the total number of people who are actually using the product and it captures things like product engagement and adoption. The next layer down is the actual body of the cake. So, team or department metrics. This is right under the actual icing and this is where leadership translates that top icing layer number into a larger set of metrics that capture what the team is focused on in the day-to-day job. So, in terms of who tracks these metrics, that would be people like directors, team leaders, and managers.

And these metrics directly contribute to the icing layer metric and represent team priorities. >> [music] >> So, for example, again, if I was a product analyst and I was working with a product team, the team level metrics would be things like retention rate or adoption rate. And retention rate is the percent of people who stick around month to month or year over year given their satisfaction with the product. The bottom layer is individual metrics.

This is where you have the base that's holding up the entire cake. And this is often the most detailed or most granular layer metrics. Think of all of the tiny grains that make up the crust of the cake. So, without it, you don't really have a cake at all. These metrics are usually tracked by employees or managers. And these are actually metrics that have more to do with productivity, capacity, and accountability. So, for example, again, if I'm a product analyst and I wanted to track the productivity of a product team or the progress that we're making as a team, then I might look at something like the numbers of features shipped.

And this is the actual changes that an individual product manager or an analyst help implement in the product. When we're thinking about metrics, we want to actually look at this cake all together, not as three separate slices. Because each of these layers actually directly feed into the layer above. In our example, the number of features shipped actually directly impacts the adoption and retention rate. And adoption and retention rate also directly impact the total number of active users.

So, if we put it all together, it looks something like this. We've got our top layer company metrics, department and team metrics, and then individual level metrics. And this is what it looks like for one vertical across product. So, when it comes to the question of what metrics actually matter, the answer is it depends on what layer of the cake you're working with. So, if you're talking to someone in leadership, let's say, a CEO or someone in exec, then that usually be focused on company level metrics.

And individual level metrics are going to be a little bit less relevant to them. But if you're talking to someone like a team lead or a manager, then individual-level metrics might be exactly what they're focused on in the day-to-day. As you can see, being a good communicator is half the job of being a good data analyst, which is why we need to keep in mind what layer of the cake our audience actually lives in. By the way, if you want to download my business metrics guide and shortcut your understanding of business metrics, so you can sound like an experienced analyst in interviews and on the job, then you can download that down below.

Okay, so now let's connect this hierarchy to what your actual day-to-day looks like as a data analyst. So, across every layer of this cake, you have different versions of teams. So, it's usually some version of marketing, sales, product, operations, customer success, or finance. And the list is going to change a little bit depending on what company you're working at. So, consulting is going to be different than health insurance, but there's usually a significant overlap.

So, if you understand team metrics and how they overlap with the layer cake, you can already shortcut your way to sounding like an experienced analyst who has real knowledge about the company and industry. So, let me show you how, starting with the three teams that are most common, which is marketing, sales, and product. So, for marketing, their main question that they're asking is how is demand for our product? So, an individual-level metric, let's say someone who's actually working on the marketing team and just tracking their progress for the month, they may look at something like posts published, right?

Then the department or team metric that they're trying to directly influence is cost acquisition or cost per lead, whereas a company-level metric is going to look at the total revenue that came directly from these marketing channels. Now, sales is going to be asking question like how many new customers are we actually closing? So, their individual metric for a salesperson would be something like the total number of calls made, whereas for a department or a team, that might be something like sales cycle length or win rate.

So, how many or what percent of calls they actually bring on board as a customer. And then the high-level company-level metric would be something like revenue, the revenue that is going to directly from a sales team or total revenue or something like ARR, uh annual recurring revenue. And then, if we add the last layer here, so product, which we already spoke about, the main question that they're asking is, are people getting real value from this?

An individual metric might be tracking something like total number of features shipped, where engineers and designers are all collaborating on making these updates to the product. And then, a department or a team is looking at something like adoption rate, retention rate, activation rate. And the top-layer company metric is something like active users. So, this is what that table looks like all together. So, once you got this pattern down from marketing, sales, product, you can start to read this table in a similar way for any other team.

Now, depending on the role that you're targeting, you want to make sure that you start to study the metrics for that specific team. That's going to help you sound like you've already worked there before you've had the actual job. This is actually the exact instinct that helped my student Nimrat, who was a student from my most recent cohort, who just landed a job as an inventory analyst at Sonoco, which is a huge fuel and logistics company.

So, when she walked into that interview, she wasn't learning this kind of language for the first time. She was already translating skills that she had from another context to this version of that company. So, a North Star metric actually measures the core value a product delivers, while also tracking long-term business growth. So, you can think of these metrics as an actionable compass. I'm going to give you an example in a second.

And then, a vanity metric looks impressive on a slide, but it doesn't actually translate to meaningful business results. So, you can think of hype metrics here. And these are easily inflatable and rarely actually guide business decisions. I remember when I was working as a data analyst at Vimeo, and during the company's IPO, I was the lead data analyst who was working directly with the investor relations team to calculate and design the metrics that we were going to share with the public.

So, we had an investor page that basically bragged about video usage. So, the total number of minutes watched, the total number of users who had ever logged in, the total number of videos ever uploaded. These are really big, exciting numbers. But those numbers don't actually tell you how the person engaged with the product in the day-to-day. So, that's what vanity metrics are for. Here's another example. So, 50 plus improvements and 100 plus bug fixes.

Without more context on what these bugs and actual improvements are, these are vanity metrics that create hype more than capture real value. So, let's go back to our product example. Some North Star metrics on the product side for Vimeo would be something like active users or average number of videos uploaded per active user, right? So, these have directly to do with how happy and engaged people are with the product.

Whereas a vanity metric might be something like the total number of bugs fixed or the total minutes watched across time. This could be something in the millions or the billions. I don't actually remember, but you can see that these two metrics can sound like they're really, really big without actually changing or impacting the direct customer value. That's why when you're working on portfolio projects or talking about examples in interviews, you want to make sure that you're focusing on North Star metrics instead of vanity metrics.

So, if you want to run a test for if something is a North Star metric for your portfolios and thing about interviews, remember that a North Star metric can only go up when the business is genuinely getting better. I think most of the time North Star metrics are those that you want to go up. There might be some edge cases where a metric is actually something that you want to decrease. But most of the time, you want to raise that number.

So, if someone can game that number or it could rise while the underlying business is actually getting worse, then it's likely a vanity metric. So, once you have these three frameworks, right? We have our metric cake, we have our North Star versus vanity metrics, and we also understand how this maps onto team metrics, we want to put this together and use it in portfolio projects, in our interviews, in our resumes to sound like an experienced analyst.

So, here are three ways to put that into practice immediately. For portfolio projects, you want to gear your project towards a specific industry or domain knowledge instead of some generic data set. So, do not use Kaggle data sets here. You want to identify two or three North Star metrics for your target company or industry and also center your project around investigating the trends and fluctuations in those metrics, not just describing the data.

So, here's an example project where we actually define what these North Star metrics are up front and we can see that these North Star metrics are really relevant to an e-commerce company or to any kind of company selling a physical product. So, this project will stand out for those kinds of roles. For dashboards, so whether you're building for a project or on the actual job, you want to make sure that you have a really clear audience in mind and you're building for their layer of the cake.

So, if you're designing for an exec, you want to prioritize top layer metrics like revenue and active users. If you're designing for a sales team to track their team performance, you want to use metrics like total calls made and win rate. Now, on the job, this is something that we would actually do in requirements gathering with stakeholders, but that's like a good benchmark to start with. So, here's an example dashboard that tracks the number of applications to a program and you can see at the top we're focusing more on team and department metrics that we would want to operationalize in the day-to-day job.

And then under that, we have metrics that get a little bit more granular that we can then use to understand what's driving those top layer metrics. So, for your resume, when you show metrics that overlap with the industry and the team that you're applying to, it really helps a hiring manager see the relevance of your work. This is not about listing every single metric that you know, but more so about showing at least two or three metrics that show that you have experience in the same day-to-day data that you'd be working with on the job.

So, here's a resume by a student who's focusing on metrics like total sales, total revenue, material and volume savings, the size of the data set, processing time, and also the size of the sales team. If you found this video helpful, make sure to subscribe and check out my other one here about metrics, KPIs, and OKRs. See you there.