This original version is a classic example of where people go wrong with illustration, especially when data is being presented. This design more than likely came from a client brief and was probably created using tools in Microsoft Word. The client obviously used arrows and boxes to highlight information to the designer. Whilst this is absolutely fine in the initial brief, it should never be replicated in the final design.
It also highlights another issue with graphs as illustrations. Graphs are simple devices used to illustrate numerical or comparative concepts. A graph takes a large amount of data and represents it graphically; the design needs to be as simple as possible. It’s obvious here that the boxes and arrows obscure the data area, with some overlapping non-related data lines. Variable sized boxes further confuse the design.
When we look at a design brief, one question we always ask is “what if this needs to be updated?” or “How can it be made future-proof?” That is, what if this graph later needs to be updated with another data line, or another? With the original design, adding a new data line would require another text box and an arrow. Where would it go? It’s not difficult to see how a design can quickly become overcomplicated and disorganised.
Typically, a 2D graph should always have clearly labelled x- and y-axes and here, whilst the x-axis is labelled, the y-axis is not. Without a y-axis label, taken in the isolation, the graph would be meaningless. Axes do not necessarily need to be gradated and there is no need here for the y-axis as it simply indicates an increase in capacity.
The x-axis gradations are clearly marked but the vertical lines extending upwards are too obtrusive in crossing the graph data lines.
In the revised version, the ugly boxed labels and arrows have been removed and key created and placed under the graph itself. Now, the data lines are clean and uninterrupted, and the graph looks simpler, cleaner, more concise and easier to understand.
The stroke of the three value lines have been reduced to match the axes stroke. The ‘Actual capacity’ line has been redrawn to show more emphasis on the increase
For the refreshed version, dotted lines, in a lighter grey are used instead: they do not detract from the data lines in the graph. Also, to give the graph a little more colour interest, three areas, in decreasing blue tints, are used to represent years 1, 2 and 3. This is purely an aesthetic choice and, in this case, further represents the passage of time. One might interpret this as the urgency for infrastructure capacity lessening over time (one might even use red, amber and green to represent this). The three colour blocks also close off the illustration, giving it a boundary which will benefit when printed on a white background.
Note that an initial version of this revision had the ‘Actual capacity’ data line starting at the graph’s origin. Because the y-axis is not gradated and only indicates an increase in capacity, the start point is pretty much arbitrary. The lines in the revised version were then shifted up in keeping with the original. The reasoning behind this is that it is likely an IT environment will already have capacity, i.e., a capacity greater than zero.