Struggling to shape your data story? Welcome to the wonderful world of frameworks!
What if you could have a road map to guide you into trimming your message while building a great data story? That’s exactly what frameworks are and I’ll share the 7 frameworks I use most often in my job as a Data Analyst!
“Oh Andrea, but using frameworks is so boring! I want to go wild and build the Harry Potter and the Goblet of Fire of data stories!”
(that’s my favourite Harry Potter book, by the way)
Look, I’m all for using creativity and building dynamic data stories with surprising twists and turns. But sometimes using a framework does a better job at conveying the story your data is telling you. As they say, sometimes less is more.
7 frameworks I use on a daily basis as a Data Analyst:
Before we go on to the frameworks, a word of advice (and kind of a warning too):
- The framework has to fit the overall message you want to tell and not the other way around. And that’s why I do quite a bit of analysis before I choose my framework to understand what kind of story the data is telling me instead of trying to fit my story into a pre-selected shape.
- Whichever is your framework of choice, it won’t tell the story on its own. You still need to add relevant data, display it well — here’s an article I wrote to help you with this — and ensure there’s a logical flow to your analysis.
So, without further ado, here are the 7 frameworks that have helped me build reports as a Data Analyst for 6 years:
1) Overviews:
I think overviews are underrated. There, I said it. They’re usually left abandoned in the standard reports that get emailed out periodically and nobody reads them. But at times it can be more about width than depth.
We live in a world with so much data available that everyone is eager to build 70-slide decks with many angles, sections, cross-analysis, etc. But I’ve seen time and time again important insight being buried under a pile of slides.
Overviews allow you to cover more ground while keeping a very solid message. Maybe, in the same way we ask ourselves “could this meeting be an email?”, we should start asking ourselves: could this 70-slide report be an overview?
2) Basics vs. above and beyond:
I recently did an analysis for a client who was performing alright, but nothing spetacular. They had improved vs. last year but were not doing quite as well as top players in their context. The challenge here was to show them that they still had a long way to improve while keeping a positive tone to the story.
For that analysis, I framed it as “you fixed the basics, well done, now it’s time to go above and beyond”. I liked how it gave the client an impression that they were on a positive path to success. The client also felt like it was quite an encouraging and positive way to say that they could still improve.
3) Deep-dive into relevant sections/regions/categories/drivers:
This is a very good framework when you’re dealing with extensive amounts of data and not all of it is relevant. Often, stakeholders want to have a general overview of performance but also want depth. Do the overview, select the key subgroups, build a thorough analysis on those and boom — another happy customer!
The only pitfall here is you need to make sure it’s very clear why you chose that particular section to go deeper into and why it’s relevant to focus on it.
4) Triggers and barriers (aka strengths and weaknesses, drivers and barriers, etc):
The bread-and-butter of frameworks and some would deem it as too cliche. I personally think this framework adds balance to a report. It gives you the opportunity to congratulate your stakeholder for good performance while also highlighting points of attention.
Don’t be afraid to use it, haters will be haters.
5) Best practices:
This is when you dissect the performance of your top store/region/category and see what the other stores/regions/categories can learn from it. This can generate a lot of good insight but it’s quite easy to travel to Utopia here and share lessons that don’t apply to the reality of the other subgroups. So, keep your feet on the ground and make sure the learnings are actionable and practical.
You can also be quite bold with this framework. Once, while I was working in the Health & Beauty industry, I sat my whole company down to learn about what our main competitor was doing well and how we could apply it to our products. It was controversial, to say the least, but if your stakeholders can look at it with mature, realistic eyes, there’re some great lessons to take away from this type of analysis!
6) Highlighting contrasts:
This framework is perhaps the lovechild of Triggers and Barriers and Best Practices. What if you took your top performing subgroup and compare it to your bottom performing subgroup?
There could be an opportunity here to really dissect their performance and understand what factors (internal or external) are contributing to their different fates. Although I recognise it can be a tough analysis to build, some really good insights can come out of it.
7) Zoom-in/zoom-out:
To be honest, I don’t use this one as often as the others because you need to have a wide scope of data to make it relevant and that’s not always the case for me. However, this framework is generally useful when the context is relevant to your story.
You may want to start small, focusing on a product and zooming out to the market context. Or you can start with market context and then focus onto a particular product. Similarly to the Deep-dive framework, make sure it’s clear why you selected that particular subgroup to zoom into or zoom out of.
I feel like having these frameworks at the back of your mind when building a story can really contribute to a great story being born. You can follow them to the teeth, you can mix-and-match them or you can use them as a stepping stone to build a more authentic analysis.
Do you also have your go-to frameworks?
