how to get the number of same users we lose from previous period to the next period

[#1]

hi,

I am trying to create a chart to measure how many same Qlikview users we lose from the previous month to the next month? The data I use is the qlikview log files.

for example:

The x-axix is month, the y axis is how many same users we lose from previous month to current month.

if

in month 1, we have 21 unique qlikview users.

in month 2, we have 40 unique qlikview users, but only 11 unique qlikview users are same as in month 1.

in month 3, we have 27 unique qlikview users, but only 20 unnique qlikview users are same as in month 2.

then

in month 2: we lose 10 unique users (=21 - 11).

in month 3, we lose 20 qunique users (=40 - 20).

I am trying to get the expression in the chart, not from the load part since I am trying to get the numbers change dynamicly when we do different selections in qlik.

Is there anyway we can create a expression for this measure dynamicly ?? I've trying different ways, but still cannot find a solution. not sure if set anaylysis could help.

could anyone help?

thank you.

[hector]

Hi, there is an example of this issue in the "what's new in qv9", and it's developed using the feature P() and E() for Set Analysis.

I'll take a look of the file during the day, maybe i can help you

rgds

[johnw]

Here's a possible solution. It shows new customers, lost customers, and retained customers for each month. It does involve script changes, but no values are precomputed, so it remains sensitive to user selections.