Top 5 things that can make better analyses
Techniques here, techniques there, techniques everywhere. With so much possible with numbers today, there’s no doubt that there is always room for improvement in any analysis. Be it your high school project or a top-level executive report in an MNC. There is always something you can add to an analysis to make it better.
But that is only when your foundational analysis is strong.
You need to have enough on the plate to choose an element out of and understand more. In this article, we’ll go over 5 things that I usually do to make better analyses. Of course, they may not be perfect but it lays the perfect foundation I need to deep-dive or answer business problems.
Feel free to skip to any section of this article that you may not have tried out before or just skim through it, we’re covering the following here;
- Understanding what is it that you’re answering?
- The left, right, center, top, down, adjacent view of things.
- The evidence and the double-check
- Understanding who you are and who is your audience
- Knowing what comes next
Understanding what is it that you’re answering?
Every analysis you do and every single number you view. It must relate to the business problem/question at hand. Knowing and understanding what you’re trying to answer will only make an analysis better.
Try and keep everything you have as close to the question as possible. Knowing that your sales have improved is good but if your question is about increasing costs, that should be what you stick to and deep dive on.
It’s better if you can find relevant underlying factors and have the perfect series of cause-effect relationships that goes back to answering your question.
If you’re finding out how your revenue is changing, understanding sales and cost is key and how they circle back to the revenue. If you try to relate it to the weather, it won’t make sense unless you’re working in agriculture or a field that is actually affected by it.
If you’re working in a coffee shop, trying to see why customers are not finishing their drinks, it won’t make sense to notice the temperature of the AC but it would make more sense to know how sweet or delightful your coffee is.
This leads me to the next thing.
The left, right, center, top, down, adjacent view of things.
Assume you own a business and the question is the same as in the previous example — How is revenue changing?
- It’s going upwards? Great! — But what is the forecast suggesting about the trend and when are we going to reach the saturation point given the current growth rate?
- It’s driven by all products? Great. — But what is the difference between the products, is the growth equally distributed across all products?
- It’s better than last month? Even better. — but how has the growth been vs the last year and how is it changing month on month this year?
- The growth rate is 5% every month? Wow. — Is this sustainable? What’s the percentage change in growth week-on-week for a more recent understanding of the numbers?
- The stock is flowing out quickly? Amazing. — But what’s the average lifetime of the stock currently in inventory is at?
I think you see my point. You should ask the right questions to always ensure you’re not missing out on any angle in your analysis.
The evidence and the double-check
The one quote I love about data — “In God and Data we trust”
There is no doubt that any analysis can be a turning point for a business. The impact that a good analysis can have on a business or organization is exponential.
When does it work in the business’s favor though? When the numbers are right. Well, the simple question then — How do you trust your numbers? How do you know they are right?
Think of it like this; if your sales are high and your costs are low, how can you still be running losses? In such cases where you know that things do look off, and to certainly avoid it, it’s imperative to run a double-check for your data and if possible, provide evidence for the same.
Adding the wrong units (millions and not thousands) or just missing out on a % symbol next to a number can give the audience a completely wrong picture of your inferences.
To avoid this, track the storyline, see how things change, and see what numbers must look like when something looks off — double-check.
Understanding who you are and who is your audience
If you’re a financial analyst, maybe you must present financials in your report and suggest to the business how they can save/make money. Your audience probably, in this case, is the CFO who takes the final call on financial decisions in the company’s board.
If you’re a business analyst, maybe it’s important to tell your audience exactly what they need to know to take better business decisions based on actionable insights. Your audience may be a specific team in the company or the executives as well.
How does knowing this helps? You can curate your analysis with the numbers that you know your audience can use to make better decisions. Maybe the CEO won’t be able to help much when it comes to reducing bad debts or increasing cash flows, he may go after the executives or team responsible for it. All you need to do in that case is to point out that the revenue is down.
No executive would want an analysis that takes days to read when you need to make a call in the next 24 hours. Keep your analyses crisp, your action points crisper, and evidence solid.
Knowing what comes next
Let’s say you answered the question well. You also kept your analysis crisp keeping your audience in mind, and double-checked your numbers with the right evidence. You also looked at every possible angle in your analysis.
Well, that brings us to the next question — What next?
Staying so close to your analysis and numbers for a while, you definitely would start noticing things that eventually need more understanding. That’s where you go next.
The other thing to pursue is a follow-up analysis of changes that may happen over time.
What do your revenue numbers look like a month from now? Why is the growth driven by only one product and not the other? Which region is driving the growth and what kind of customers do you have there?
These are only a few things to start with but it’s always good to deep-dive on an analysis, finding the root cause, and working on things that will help the business make more informed decisions. There is always scope for growth there, and if that is not possible, move to see how things change over time.
Leave a comment if you think I missed out on any other pointers that are relevant to the article! (Thanks!)
Do leave a clap and check out my previous articles here, if they interest you!
For all my articles:
Connect with me on LinkedIn: https://www.linkedin.com/in/yash-gupta-dss/
~ P.S. All the views mentioned in the article are my sole opinions. I enjoy sharing my perspectives on Data Science. Do contact me on LinkedIn at — Yash Gupta — if you want to discuss all things related to data further!
