Key Influencers visualization - likelihood figure doesn't match

[PJ-292]

Hi,

In the Key Influencers visualization, shouldn't the likelihood figure be the difference between the selected inluencer and the average of all other factors?

E.g. the top influencer is Category A, with average weekly sales of $1860. Whilst the average weekly sales of all other category is $870.73. 

So the difference is $989.27. So shouldn't it be that: If product is in Category A, the average weekly sales is $989.27 higher than all other category?

Why is the graph showing $9.60k instead?

Kind regards,

PJ

[v-frfei-msft]

Hi @PJ-292 ,

Kindly share your sample data and excepted result to me if you don't have any Confidential Information. Please upload your files to One Drive and share the link here.

[PJ-292]

Hi @v-frfei-msft ,

I've re-created a masked data .pbix (note: the numbers are slightly different than my original post due to rounding issue, but the main problem is still there).

https://1drv.ms/u/s!AuKPBCJSdtVKvmrM3-OdWij0afT5?e=yuXnYT

What I expected to see is that:

"Weekly Sales is more likely to increase by $989.22, when Category is A than otherwise (on average)."

From the Microsoft article, it says that "The number in the bubble is still the difference between the red dotted line and green bar but it’s expressed as a number".

In my data:

Influencer (blue bar) = Category A = $1859.95

Average of all other category (red dotted line) = $870.73

$1859.95 - $870.73 = $989.22

So, it should be: When it's Category A, the average Weekly Sales increase by $989.22. I'm not sure why the chart is showing it as $9.09k.

Kind regards,

PJ

[Larsen]

Hi! 

I had the same problem, and I'm pretty sure the issue is that 0/zeros are not included in the average calculation for the influencers. So would be really nice if this is fixed asap, it does not seem like a user-issue but a design issue:)

KR Larsen