Let the data lead you, but don’t distort it.  Creators of good, strong narratives are careful about what they include as well as what they exclude — you can’t cram it all in. You have to manipulate your data, even through omission of all kinds of potentially related data. Otherwise your story gets crowded and incomprehensible. This means you’ll need to be able to find a line between manipulation and distortion, and don’t cross it. 

When John Snow, the doctor who famously discovered the contaminated well that caused London’s 1854 cholera outbreak, was representing the data he’d collected to show that Soho’s Broad Street pump was the one killing people, he didn’t just have a “eureka” moment. He was already a proponent of germ theory, and he had a good idea of the story he needed to tell, but it was the spatial representation of his research that helped convince the authorities there was something in his argument.

His story has such powerful momentum because public health was riding on it.  Your graphics might not mean the difference between life and death — so how will you make your data matter?

Test your assumptions at each step. If your story starts to fall apart, you may have to go back to the beginning. Put your story first, and let your command of the data lead you, not your desire for a particular outcome. Is there something in your analysis that needs to be rethought, or is the problem with the data? Or is the problem in your initial question?

You want to guide your audience toward your interpretation, but a great data story can leave things open to multiple readings. If you’ve never read Darrell Huff’s 1954 How to Lie With Statistics, you can read it here.  It might be 60 years old, but it’s still worth reading — the size of our datasets has changed, but the pitfalls haven’t.

Take Coco Chanel’s advice.  She advised that before you leave your house, you should look in the mirror and remove one accessory. Try this with your infographics — do you really need all those variables?

Do you have any elements that are just a distraction? Your variables are your story’s characters, and you give them life when you bring them into contact — or conflict — with one another. You need enough variables to tell a rich and compelling story that the user can be involved in. Too few variables, and your story might not say anything new; too many, and you’ll end up with a confusing set of noisy bangles.

Telling a story as simply as possible doesn’t mean that the ideal infographic is a bar chart. It’s OK for your graphic to be as busy as a Richard Scarry book, so long as you do what Scarry did: make everything count, know what excites your audience, and focus on the message, whether that message is that a contaminated water pump is killing people, or it’s that you, too, can grow up to be a cat dressed as a firefighter.

Your infographics are only as good as your story, and your data is only as useful as the value of the insights you get from it. You don’t have to be a power user to make a powerful point, but you need Excel skills to analyse your data, and organise your spreadsheets.A good, clean dataset that’s got all its cells in a row is pretty damn vital when it comes to making an infographic. We asked Aidan Corbett, CEO of Virtuoso  [Update! Virtuoso is now Kubicle] to give us some tips on effective data manipulation in Excel, and he’s great, so he did.

Tip 1: Arrange your data records in rows, not columns
First, the basics. Excel expects data records to be arranged in rows, not columns. Following this convention will make sorting, formatting and filtering data much easier. The first row of cells should contain the headings of each column (e.g. Date, Address, Email address). Below this, arrange one data record on each row.  Don’t worry about running out of rows — there are more than a million of them available on each sheet.
​
Tip 2: Add and delete rows, not cells.
If you want to add or delete data records, the safest way to do this is to add or delete full rows. Moving cells or groups of cells can easily create errors in your dataset. This is the  kind of dead reckoning mistake that can skew your whole analysis. Don’t do it! Move the whole row (or save the original dataset separately, which is a good idea anyway)  if you don’t want to lose the content.


Tip 3: Use Sort and Filter to search for relevant data.
Sort and Filter allow you to quickly search for data using dropdown buttons on each column. First select the full dataset (including the headings), then go to the ‘Data’ tab in the ribbon and click on the ‘Filter’ option. To remove filters, just press the ‘Clear’ button button in the same tab.

​
Tip 4: Change the format of your data.
Sometimes data columns are not presented in the ideal format. For example, I might want to add a $ sign and comma separators to a revenue column. To make changes to your data format, select the column of data to be changed and then press the arrow button in the bottom righthand corner of the ‘Number’ section of the ‘Home’ tab. This displays the Format Cell dialog box, which will allow you format your column in a myriad of ways.

Tip 5: Remove the gridlines.
I find that removing the gridlines on my worksheet often makes data easier to read. To do this, simply go to the View tab on the Ribbon and uncheck the ‘Gridlines’ box in the ‘Show’ section of the ribbon.

​Tip 6: Use different colours for formulas and raw data.
Some datasets combine raw data with formulas (e.g. Quantity x Price = Total revenue). If your dataset contains columns of formulas, I recommend changing the colour of these numbers to distinguish them from the raw data.
​
Tip 7: Format your data as a Table
Excel allows you to format your data as a Table. Tables add great format styles to your dataset and automatically include filter buttons on top of each column. Simply select the full dataset (including the headings) and click the ‘Format as Table’ button in the ‘Home’ tab of the ribbon.

Two weeks ago, the creator and star of Veronica Mars announced that they were crowdfunding a movie based on the cult hit series from the early 2000s. Rob Thomas set a Kickstarter goal of $2million to get the project made, and they made Kickstarter history by getting to their funding goal within ten hours. And it just kept going. At the time of writing, backers have pledged $4,190,613. (UPDATE: I’ve even revised that upward since I started writing, in an attempt to conceal just how slow a writer I am. UPDATE: OK, twice now.)

It worried everyone from entertainment lawyers to pop culture critics, for reasons ranging from self-promotion-inspired sour grapes to legitimate concerns about the quality of storytelling that might result from a fan-driven campaign. And we wondered if the whole funding pattern looked a little uneven, with a focus that too heavily privileged the already privileged. Last year, Amanda Palmer raised more than a million dollars after looking for $100k to complete her record, book, and tour. What’s with all the overkill? (But this also comes in the wake of a study showing that Kickstarter’s pledges may be slowing down.)

We worried a little if it meant that smaller projects might get crowded out by the big name stars. So we had a closer look at what gets funded, and the results surprised us. We’re used to backing friends’ crowdfunded albums, films, and other audience-focused projects, so what we didn’t realise is that the projects that get the most funding are those in design  and games. In fact, if you look at the numbers, design and games are particularly likely to end up overshooting their funding targets.

But then last week, a Maryland third-grader wanted to raise $829 to prove to her older brothers that girls could make games. At this moment, tiny fashionista and our new hero, Mackenzie Wilson, has raised more than $22,000. She said that she’d been inspired by the success of the Veronica Mars campaign. And in any case, we learned that it’s projects like Kenzie’s that get the most funding. Maybe we were a little too quick to judge.