A year ago I wrote a rather long post about the unexpected fact that books that have recently arrived in store sell much better than books that have been around a while.

Turns out it is still true.

This chart shows the proportion of each month's batch sold after six months (in blue) and 12 months (in orange). The series starts in August 2017 and runs until March 2019. We don't have 12 months data after September 2018 yet, which is why the orange bars stop then.

You can see what I can see, which is that:

After six months, we have usually sold 50 to 60 per cent of each batch (the blue bars). Our best ever (March 2019) is 67 per cent. Our worst (August 2017) just 35 per cent.
In the following six months, we sell about another 15 per cent or so of each batch (the orange bars). After 12 months, we have usually sold between 60 and 70 per cent.
The proportion sold after six months seems to be rising over time, which might reflect the fact that gradually sales are improving.

The proportion sold does not seem to be related to the number processed. The most we ever processed in a month was in June 2018. We sold 50 per cent within six months. In May 2018, the month before, we processed only 325. We had also sold 50 per cent of these six months later. This chart shows the number processed by month over time.

Trying

In my earlier post, I said we'd do an experiment to try to spot an impact on overall sales if we processed a very large number of books. The month we chose was November 2018 and I wanted to process 2,000. As it turned out, we processed 1,507 with the books that we could source and the time that we had to devote to this task amongst all the other tasks.

You can see the month on the previous chart, the fifth from the right. We sold 58 per cent of these books within six months. This may not seem like a good result, ie we processed all those extra books but still sold the same proportion as usual. But from a revenue point of view, selling 58 per cent of 1,500 books is twice as good as selling 58 per cent of 750 books. So I think we can conclude something important from this: processing more books generates us more revenue.