Compliance reports don't tell stories

Many annual reports are written to satisfy legal requirements. They are a type of compliance report created for accountability. There's nothing wrong with this kind of reporting—it needs to exist. But publishing information doesn't automatically mean it is understood.

Let's look at an example. Wellington City Council's (WCC) 2024/25 annual report is hundreds of pages long and includes a lot of financial data. The council must produce it each year under the Local Government Act 2002.

The numbers presented in this report are accurate and accessible, but as you read through it, they're rarely explained. I question how effective this is for the everyday citizen, or for someone not working for the council, or even for someone without a background in finance.

So, here's my process of taking this compliance report and telling one of the many data stories it hides.

Numbers inform (but only if you already understand them)

Page 4 of the WCC report shows a snapshot of important metrics. This type of data visual is called a BAN (big ass number), and yes, that's the technical term. BANs are often used because they make the data look a little more accessible—one number is easy enough to understand, right?

We know from one of these BANs that the council made a surplus of $61.7m. Another BAN shows that their capital expenditure was $488.6m, an underspend of 30% on the budget. Net borrowing was $1.6b, up from $1.4b the year before. And only 58% of performance targets were met.

Wellington City Council Annual Report FY 24/25 (page 4)

You're forgiven for being confused by these numbers—I was. A surplus sounds positive. An underspend sounds prudent. But debt grew despite this, and nearly half of the targets weren't met. How is this even possible?

You may have been able to join the dots between these BANs and create your own understanding. Or you (like me when I first read these numbers) may not have knowledge of council operations to do this.

If you're in the second group, more numbers won't help you understand this data. What you need is a data story.

Data stories educate

Data stories explain the meaning of data. They're written for an audience who are missing the context they need to join those dots themselves. So, before you write a data story, you'll need to find this for them. Every data metric comes with its own context. Here's how I found this for the WCC BAN, Capital Expenditure.

To make the $488.6m number relative, I'm going to compare it to the budgeted expenditure of $695m. This gives us an underspend of 30%.

Here are five questions you can then ask about any data metric to understand its wider context:

• What happened?

• Why does it matter?

• How did it happen?

• What don't we know?

• What does this mean for the future?

Capital Expenditure Answers

For the capital underspend, WCC spent 30% less than planned on capital projects, including key infrastructure. This matters because Wellington's infrastructure is ageing, and most of it will need to be replaced within 30 years.

The underspend happened for two main reasons:

1. The council cut back its infrastructure programme after deciding not to sell its airport shares. In the event of a future natural disaster, it had to free up borrowing capacity to build financial resilience.

   
   

2. The central government cut its transport funding.

I don't know exactly how much of the underspend was deliberate versus unavoidable, or how much was absorbed by cost inflation, because this wasn't detailed in the report. Remember, there will always be things you don't know about your data and this can sometimes form part of your story about it.

Looking to the future, the deferred work still needs to be done, and it will likely cost more.

I also went through the five questions for Net Debt. I haven't included those answers here, but the process is the same. Answering these questions for each data metric that you want to explain will help you communicate it to an audience who can't understand the numbers alone.

The next step is to arrange the information into a story.

Stories have three acts

Stories have three acts, and data stories are no exception. Act 1 sets the scene, Act 2 introduces the conflict, and Act 3 resolves it.

So, with a focus on the Capital Expenditure and Net Debt BANs, I structure the context from my questions into three acts.

Summary Data Story

Act 1: Wellington's infrastructure is ageing, and most of it will need to be replaced within 30 years. As a city built on a major fault line, Wellington also needs to keep financial reserves available to respond to a future disaster (rather than committing it all to planned infrastructure).

Act 2: But rates and borrowing—the two main ways to fund both—are close to their limits. Other funding alternatives come with their own cost or limitations.

Act 3: Therefore, the Council is in a difficult position: the infrastructure clock is ticking, and the disaster risk is real. Understanding how Wellington pays for its future is one of the most important financial questions facing the city right now.

You might notice that apart from the "30 years", there are no other numbers in this story. A data story is not how you share data—it's how you share the MESSAGE of data.

This is an example of a summary data story. It's high-level and what's necessary to educate a novice audience, without overwhelming them with detail.

But what if I need them to understand more detail?

Then, I create what's known as a detail data story. I pick a part of the summary story and write another story (think, three acts) to help expand it.

For this story, I want to add detail to this line from the second act: Other funding alternatives come with their own cost or limitations.

Detail Data Story

Act 1: One way the council has tried to manage the funding constraint is to cut back its infrastructure programme. In FY 24/25, Wellington City Council spent 30% less than planned on capital projects. Part of this was a deliberate decision to build financial resilience. Part of it was unavoidable—the central government cut its transport funding. This reflects an ongoing trend of underspending.

Act 2: But despite the smaller spend, the council’s debt still grew by $207m because existing borrowing commitments did not stop when the work did.

Act 3: Therefore, cutting the infrastructure programme is not a sustainable alternative. The work still needs to happen, and deferring it usually makes it more expensive.

I can now use these two related data stories to help people understand the data metrics. My next step is to think about how I'll tell them.

Data visuals help tell data stories

It's important to realise that data visuals aren't data stories. A data story is a narrative. It's what you've just read. See Data storytelling vs data visualisation: what's the difference?

However, data visuals can be very helpful for some audiences in understanding data stories. If you choose to use data visuals to tell your data story, let the story drive what data visuals you create. Ask, "What parts of this story would be better told using a data visual?" This will give you some pretty good graph titles.

I decide there are no parts of my summary data story that a data visual could help tell, and I'm not going to create something for the sake of it, so I move on to my detail data story. I can produce two data visuals from this to help communicate it.

Here's the first.

Visual to support Act 1 of the detail data story

The visual is designed to communicate its title, which was taken from the data story. The title is about capital expenditure, so that's what's graphed. The title focuses on FY 24/25, so I made that the focus (blue). It also mentions a trend, so I show several years of data. Every design decision serves the telling of the data story.

Next up, is one of the more complex visuals I created. A waterfall graph may not suit every audience. Knowing your audience's data visual literacy helps you design something they'll understand. If you're unsure, stick to the basics—bar, line, pie (yes, pie!).

Visual to support Act 2 of the detail data story

There are many ways I could tell my full data story. I could display the visuals on slides during a presentation. I could create a report. I could publish it on a blog. I could turn it into a video update (shoutout to Andrew Little and Ben McNulty, Wellington's Mayor and Deputy Mayor, for their weekly social media updates). I could also display it on a one-pager like the one below.

The summary (highlighted) and detailed data story, including visuals

How you share your data story should align with your audience's preferences. If you have a large, diverse audience, you'll likely need to repackage the same story differently.

My Advice For WCC

I'm in no way suggesting WCC should get rid of its annual report. As I said in the beginning, this kind of compliance reporting serves a purpose.

What I think the council should consider is how else they communicate this information. And maybe they're already doing this. I don't know.

I'd like to imagine they have lunch-and-learn sessions where things like this are explained, at least to the people who work with and for the council. It would be amazing if this information could also be explained to the people of Wellington.

It took me much longer than it should have to understand what seems like a relatively simple story about two data metrics. The majority of people wouldn't have bothered—and that's the problem with annual reports.

If you work with compliance reports, the data stories are already there. Your job is to present them with the appropriate context for your audience.

Kat Greenbrook is a data storytelling consultant, author, and workshop facilitator based in Wellington, New Zealand. She is the founder of Rogue Penguin and the author of The Data Storyteller's Handbook.