Reuse Custom Metrics in DQL Without Reinitialisation

knallapa — Wed, 17 Dec 2025 13:44:03 GMT

I'm currently using this query to reuse the same custom metric at different time intervals to populate table data in my dashboard. However, I’d like to simplify the query and reuse a variable multiple times with different conditions in DQL. Is there a better way to structure this?

timeseries METRIC_A_approved = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeA" AND
(responseCode == "0" OR responseCode == "00") AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }

| join [
timeseries METRIC_A_total = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeA" AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_A_total }, kind: outer

| join [
timeseries METRIC_B_approved = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
(responseCode == "0" OR responseCode == "00") AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_approved }, kind: outer

| join [
timeseries METRIC_B_total = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_total }, kind: outer

| join [
timeseries METRIC_B_approved_subgroup = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
(responseCode == "0" OR responseCode == "00") AND
(categoryCode in ["1001", "1002", "1003", "1004", "1005"]) AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_approved_subgroup }, kind: outer

| join [
timeseries METRIC_B_total_subgroup = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
(categoryCode in ["1001", "1002", "1003", "1004", "1005"]) AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_total_subgroup }, kind: outer

| join [
timeseries METRIC_B_approved_domestic = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
(responseCode == "0" OR responseCode == "00") AND
categoryCode == "2001" AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_approved_domestic }, kind: outer

| join [
timeseries METRIC_B_total_domestic = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
categoryCode == "2001" AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_total_domestic }, kind: outer

| join [
timeseries METRIC_B_approved_crossborder = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
(responseCode == "0" OR responseCode == "00") AND
categoryCode == "2002" AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_approved_crossborder }, kind: outer

| join [
timeseries METRIC_B_total_crossborder = sum(
`custom.app-metrics.transaction.data`,
scalar: true,
filter: {
txType == "TypeB" AND
categoryCode == "2002" AND
(isNull($Entity) OR $Entity == "" OR entity == $Entity)
}
), by: { entity }
], on: { entity }, fields: { METRIC_B_total_crossborder }, kind: outer

| fieldsAdd
Rate_A = if(coalesce(METRIC_A_total, 0) == 0, 0, 100 * coalesce(METRIC_A_approved, 0) / coalesce(METRIC_A_total, 0)),
Rate_B = if(coalesce(METRIC_B_total, 0) == 0, 0, 100 * coalesce(METRIC_B_approved, 0) / coalesce(METRIC_B_total, 0)),
Rate_B_subgroup = if(coalesce(METRIC_B_total_subgroup, 0) == 0, 0, 100 * coalesce(METRIC_B_approved_subgroup, 0) / coalesce(METRIC_B_total_subgroup, 0)),
Rate_B_domestic = if(coalesce(METRIC_B_total_domestic, 0) == 0, 0, 100 * coalesce(METRIC_B_approved_domestic, 0) / coalesce(METRIC_B_total_domestic, 0)),
Rate_B_crossborder = if(coalesce(METRIC_B_total_crossborder, 0) == 0, 0, 100 * coalesce(METRIC_B_approved_crossborder, 0) / coalesce(METRIC_B_total_crossborder, 0))

| filter not(isNull(entity) or entity == "")
| fieldsAdd
Entity = entity,
A.Total = toLong(coalesce(METRIC_A_total, 0)),
A.Rate = concat(toString(round(coalesce(Rate_A, 0), decimals: 2)), "%"),
B.Total = toLong(coalesce(METRIC_B_total, 0)),
B.Rate = concat(toString(round(coalesce(Rate_B, 0), decimals: 2)), "%"),
B.Total.Sub = toLong(coalesce(METRIC_B_total_subgroup, 0)),
B.Rate.Sub = concat(toString(round(coalesce(Rate_B_subgroup, 0), decimals: 2)), "%"),
B.Total.Dom = toLong(coalesce(METRIC_B_total_domestic, 0)),
B.Rate.Dom = concat(toString(round(coalesce(Rate_B_domestic, 0), decimals: 2)), "%"),
B.Total.XB = toLong(coalesce(METRIC_B_total_crossborder, 0)),
B.Rate.XB = concat(toString(round(coalesce(Rate_B_crossborder, 0), decimals: 2)), "%")

| fields Entity, A.Total, A.Rate, B.Total, B.Rate, B.Total.Dom, B.Rate.Dom, B.Total.XB, B.Rate.XB, B.Total.Sub, B.Rate.Sub
| sort A.Rate desc

Re: How can I reuse a custom metric in DQL without reinitializing it multiple times?

Maheedhar_T — Thu, 24 Jul 2025 19:23:13 GMT

Hi @knallapa ,
You can set up the common thing in all queries as variable and reference it in dashboard tiles.
Reference: Add a variable to a dashboard — Dynatrace Docs
You can add as many variables as you want to. Maybe set up one variable for totalTransactions then use that in calculations. Set up a constant timeframe for that variable and let the other tiles have dashboard timeframe.

Regards,
@Maheedhar_T

Re: How can I reuse a custom metric in DQL without reinitializing it multiple times?

knallapa — Thu, 24 Jul 2025 19:31:20 GMT

Sure, I will try in that direction. Thankyou @Maheedhar_T

Re: How can I reuse a custom metric in DQL without reinitializing it multiple times?

knallapa — Thu, 31 Jul 2025 04:24:12 GMT

For Above Query Nothing worked for me: Tried variables, Problems i faced is :

can’t fetch a metric once and reuse it with conditional logic applying filters.
can’t assign a timeseries result to a variable inside fetch.
can’t use let to define reusable filters inside fetch.

Moreover variables have limits. other issue was variables i can't extend and apply filter conditions to that variable.

topic Re: How can I reuse a custom metric in DQL without reinitializing it multiple times? in DQL

Reuse Custom Metrics in DQL Without Reinitialisation

Re: How can I reuse a custom metric in DQL without reinitializing it multiple times?

Re: How can I reuse a custom metric in DQL without reinitializing it multiple times?

Re: How can I reuse a custom metric in DQL without reinitializing it multiple times?