Four Things I Wish Every Chart Did

Main illustration: Katie Garth

Analytics tools are great for collecting data that’s easy to measure, and visualising it in beautiful charts. Sadly, this usually leaves you with more questions than answers.

Unlike other tools, analytics products are primarily read-only. You don’t go there to take action, you go there to get answers and insights so that you know to take action elsewhere.

These products should focus on the job of “answering questions and providing insights” for their customers, rather than falling into the category moat of offering the stock line/bar/pie chart with a time selector; that’s what everyone else is selling.

When traffic spikes, the question is Why? When conversions drop, the question is Why? When sign-ups flatline, the question is Why? Leaving your customers to do visual gymnastics as they try to piece together parts of your product is a bad experience.

Here are four improvements that’ll make your visualisations more about answers, and less about gradients and drop shadows

1. Annotate The Data

Any time there’s a large spike or dip, it’s usually related to a change elsewhere in the system. Either a new referrer has appeared, an old one has disappeared, traffic from a different page has increased, or there’s been a decline in conversions from the previous step. A simple improvement would be to show the likely cause with the effect.

2. Annotate The Slopes

Most of the time wasted chart-gazing is spent eyeballing lines to work out their relative slopes and inclines. How quickly are we growing? Are we growing faster than before? etc.

It’s never the point that matters, it’s the trajectory of the line. No one cares if you have $1MM in MRR, if you’re dropping 10% month over month. Once again analytics tools make this the harder piece to find, favouring irrelevant precision, rather than meaningful insight.

3. Exclude (or flag) incomplete periods

Every chart gets this wrong. For some reasons, unless it’s midnight on a Sunday every weekly chart I look at collapses at the final period. Unless its the 31st of the month, every monthly chart drops off a cliff too. Technically it’s “correct”, but it’s also dumb. It doesn’t aid analysis, and confuses more often than clarifies.

A simple improvement would be to either flag incomplete periods, or, better yet, show what they’re trending towards for that period. I realise that sounds crazy, but maybe it’s time analytics tools start doing some analysis.

4. Enable projections

Sticking with the theme of analysis, trajectories always lead to projections. Tell a company that their revenue is growing 10% month-on-month and their next question will be along the lines of “How long until we hit $100K?” or “What will it be like in August?”. Sadly the best and often only way to do that these days is to take the data to Excel and run some formulae, some products don’t even make the export easy.

Focus on usage, not categories or competitors

If you focus the product on what people are actually using it for, these ideas start to become obvious. If the next step is a Google Doc, then integrate with Google Docs. If the next step is Excel, then make export to xls easy. If the next step is to create a Keynote slide, then make it easy to get a high resolution image without any UI present. Support your users in what they’re doing with your product.

If you obsess about the jobs your users are doing, you’re set up to improve your product in ways that they will value. If you obsess over your category or your competitors you’ll end up offering 3D bar charts and multi-gradient visuals, because that’s what everyone else has. Your feature checklist will be complete, but your users will hold rulers against their monitors trying to get the answers they’re paying you for.