In France we usually say that « a good drawing is better than a long explanation« .
This is only true to some extent.
In the world of Web Analytics, the very nature source of graphs is structured data, if you go beyond the 5-column bar chart, that is 😉
The problem behind structured data is that we tend to organize it in tree structures. Which branch out. Over and over.
Soon, your tree diagram will turn out to have either very tall/wide foliage or deep/extensive roots. Either way, your graph becomes unpractical outside of your personal screen.
This is usually due to the fact that:
- You can’t pan left/right/up or down enough to get to the information you want right away.
- Zooming in or out leaves you either missing out on the big picture or being too far away that you miss all the important bits.
- You cannot present your findings because it won’t fit into a Word document or a Powerpoint slideset, as explained in the above points (Omniture Next Page Flow report, anyone?).
In other words, when the information intake is just too important or just un-workable, you have to find ways of making that information more practical.
Enter Flash graphs and drill-down listings! While not the best solution out there, Flash has the merit of being an Adobe product. The company who brought you Photoshop has very strong ties and influence with everything related to the SVG XML-based graphics format. If your reporting data is structured enough, it can be formatted using XML and then processed by Flash to render your graph… with the addition of functionalities such as drilling down, collapsable containers, and in-set zooming tools.
Don’t get me wrong, I do not own Adobe stock but you have to admit that using any reporting tool that uses Flash graphs to describe a huge amount of data is usually a more powerful reporting experience than with just an image. That and it’s usually a great toy in an even better reporting toolbox!
WA companies such as Google Analytics and Visual Sciences understood that notion.
For reporting on smaller amounts of data, go with regular reports; heck, use images!… as long as it’s better than a long explanation 😉