USQL → DQL for RUM Data

USQL → DQL for RUM Data

Gen2 used USQL (User Session Query Language) to analyze Real User Monitoring data. Gen3 replaces it entirely with DQL using fetch user.events and fetch user.sessions.

Side-by-Side

USQL (Gen2)                             DQL (Gen3)
──────────────────────────────────────  ──────────────────────────────────────
SELECT browserType, COUNT(*)            fetch user.events
FROM usersession                        | filter characteristics.has_page_summary == true
GROUP BY browserType                    | summarize count(), by:{browser.name}

SELECT AVG(duration)                    fetch user.events
FROM useraction                         | filter characteristics.has_navigation == true
WHERE application = "MyApp"             | filter frontend.name == "MyApp"
                                        | summarize avg_duration = avg(duration)

SELECT TOP 10 name, AVG(duration)       fetch user.events
FROM useraction                         | filter characteristics.has_page_summary == true
GROUP BY name                           | summarize avg_dur = avg(duration), by:{page.url.path}
ORDER BY AVG(duration) DESC             | sort avg_dur desc
                                        | limit 10

Data Source Mapping

USQL Table                              DQL Data Object
──────────────────────────────────────  ──────────────────────────────────────
usersession                             fetch user.sessions
useraction                              fetch user.events (+ characteristics filter)
userevent                               fetch user.events

Key Field Differences

USQL Field                              DQL Field
──────────────────────────────────────  ──────────────────────────────────────
browserType                             browser.name
osFamily                                os.name
city, country                           geo.country.iso_code
duration                                duration (nanoseconds in sessions!)
application                             frontend.name
userSessionId                           dt.rum.session.id
userId                                  dt.rum.user_tag
bounce                                  characteristics.is_bounce

Important: Session Time Windows

fetch user.sessions only returns sessions that started in the query window — not sessions active during it. Sessions can last 8+ hours. For correlation queries, extend lookback by at least 8 hours.

Session Field Naming

Session fields use underscores, not dots:

✅ navigation_count          ❌ navigation.count
✅ user_interaction_count    ❌ user_interaction.count
✅ request_count             ❌ request.count
✅ page_summary_count        ❌ page_summary.count

Bounce Rate Example

fetch user.sessions, from:now() - 24h
| filter dt.rum.user_type == "real_user"
| summarize
    total = count(),
    bounces = countIf(characteristics.is_bounce == true),
    avg_duration_s = avg(toLong(duration)) / 1000000000
| fieldsAdd bounce_rate = round((bounces * 100.0) / total, decimals: 1)