I was sent this last week and overall it's not too bad:

Clearly decipherable and does show what it sets out to. However, there are some nuances that could be tweaked to make things that much easier to read.
  • Firstly, the series (L1, L2 and L2+) are denoted by a legend on the right hand side - and that legend has each series as a different colour, with the series markers being different. The reader is forced to read the legend and refer back to the graph - some decoding is necessary
  • The x-axis label is superfluous - as the reader can see the dates
  • The y-axis is misleading - the chart does not show % attainment, but rather the % of the cohort attaining each of the criteria denoted by the series
  • Gridlines
  • The chart has a coloured (pink / salmon) background
  • The title is in red
Some Excel-fu later:
  1. Legend replaced with data endpoint labels
  2. Y-axis removed
  3. Data labels used to embed the values onto the chart
  4. Each series, the same (grey) colour, with consistent marker shapes
  5. X-axis title removed
  6. Chart title changed to reflect the data displayed (and in black)

Why this is "better"
  • Reduced colour load reproduces better (photocopying) and is more accessible to people with reduced / impaired vision of colour-blindness
  • As the data does not overlap, replacing a legend with data end labels makes the linking of the chart data to the label more intuitive - you don't need to decode the chart as much
  • Removing the superfluous information reduces the noise on the chart - making the data the main event, not the colour coding / formatting.
  • At the end of the day, the "story" of this chart, is the year on year trend which by reducing all the unnecessary clutter on the chart, the lay reader is more able to decode without inside knowledge of how to read charts.