How to build a cumulative measure without date context?

[MaxViana]

I need a cumulative measure to build a Pareto chart, but my cumulative measure fails when my main measure has repeated values.

Look the problem in this dashboard:
https://app.powerbi.com/view?r=eyJrIjoiMjZlN2Q2MGYtZWYyOS00NGNlLWE3ZDYtYjQwZWIxOWUzMjMxIiwidCI6IjgxZ...

Sample:
https://1drv.ms/u/s!AoIveuLXWhhKgvFK_v7uRBhk38hPXQ?e=dP80e0

[lbendlin]

Your repeated values will not allow you to properly rank them, and will potentially prevent you from identifying the 80/20 buckets.  By adding an index column you make each row of your data unique and sortable.  Then one of the duplicate values will fall into the 80 bucket and the other in the 20 bucket.

[lbendlin]

Add an index column to your data to be able to tell these data points apart.

[MaxViana]

I think I do not understand. Could you explain a little more to me?

[v-rzhou-msft]

Hi @MaxViana 

Your problem is caused by the repeated values in Acc(It will cause the repeated rank).

So when you use top N, all "Município Fato" with same rank will get the same cumulative result.

So it's is better for you to get another rank for "Município Fato" in same rank.

You see 'Paraíso do Tocantins' and 'Pedro Afonso' have the same ranking, try to add a new rank(ranking2)  'Paraíso do Tocantins' = 1,  'Pedro Afonso' = 2 .

Then calculate a new value = ranking + ranking2/100.  'Paraíso do Tocantins' = 4.01 and 'Pedro Afonso' = 4.02.

Finally you calculate a new rank by this value, you will get unique ranks.

Best Regards,

Rico Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.