![Manga D.](/assets/transparent-ad5be28fbcd25b7b08d2cebe1d957125437fb5407d75ee717965ad22c8808791.gif "Manga D.")
MD

Manga D.

Data Engineer

Mid-Market (51-1000 emp.)

6/25/2026

"Automated Monitoring and Lineage That Quickly Boost Data Trust"

4/5

What do you like best about Monte Carlo?

The biggest value for us has been Monte Carlo's automated monitors. Instead of hand-writing freshness and volume checks for hundreds of Snowflake tables, the ML-based detectors learn normal patterns and alert us on anomalies automatically — this caught a stalled pipeline load hours before our business stakeholders would have, and saved us from reporting on stale numbers.

The dbt and Snowflake integrations were quick to connect and are a core part of our daily workflow. End-to-end lineage is the feature I rely on most: when an alert fires, I can trace it from the downstream table back through the dbt models to the exact upstream source in a couple of clicks, which has cut our root-cause investigation time from hours to minutes.

On UI/UX, the incident view and Slack alerting keep the whole data team in the loop without anyone having to log in and dig around — alerts land in our channels with enough context to triage right away. Performance has been solid even across our larger warehouses, and the monitors run without us having to manage any extra infrastructure.

In terms of ROI, the time we save on building/maintaining custom data quality checks and on faster incident resolution has easily justified the cost. Onboarding and support were smooth — the team helped us get our key tables monitored quickly, and an unexpected benefit has been how the lineage and monitoring have improved data trust across the org, so stakeholders rely on the data more and we field fewer "is this number right?" questions. Review collected by and hosted on G2.com.

What do you dislike about Monte Carlo?

The biggest pain point for us is pricing and credit consumption. Some features, like certain monitors and the PR/CI integrations, burn credits in ways that aren’t always clear up front. Because of that, we’ve had to regularly review what’s actually being used and disable integrations we rarely rely on just to keep costs in check. Clearer, more predictable visibility into per-feature costs would help a lot.

The automated monitors can also be noisy at first. During the initial learning period, we saw a fair number of false-positive alerts, which meant manual tuning and some effort to set sensible thresholds before the signal-to-noise ratio improved.

On the UI/UX side, moving between lineage, monitors, and incident details can take a lot of clicks. The interface also has a bit of a learning curve for newer team members, especially those who don’t use it every day.

Finally, custom/SQL-based monitors are powerful, but they’re not as intuitive to set up as the out-of-the-box options. Getting solid coverage for sources outside the main warehouse, versus our core Snowflake/dbt tables, also takes more effort. None of these are dealbreakers, but they’re the areas where we’d most like to see improvement. Review collected by and hosted on G2.com.

What problems is Monte Carlo solving and how is that benefiting you?

Monte Carlo has helped us solve a real data quality and observability gap. Before adopting it, we had limited visibility into the health of our Snowflake and dbt pipelines. Problems like stale tables, failed loads, volume drops, or unexpected schema changes could easily slip by and only surface when a stakeholder noticed a wrong number in a dashboard. As a result, we were stuck in reactive firefighting mode and constantly answering variations of, “Is this data correct?”

With Monte Carlo’s automated monitoring, we now catch many of these issues proactively, often before they reach downstream consumers. The upside is twofold: we spend far less time building and maintaining custom data quality checks, and we resolve incidents much faster. The end-to-end lineage is a big part of that, because it lets us trace a problem from a downstream table back to the source in minutes rather than hours.

It’s also addressed a broader data trust issue. With monitoring and lineage in place, plus alerts flowing into Slack, stakeholders have noticeably more confidence in the data, and our team gets far fewer ad-hoc “can you verify this?” requests. Overall, it’s shifted us from reactive to proactive and freed up engineering time for higher-value work. Review collected by and hosted on G2.com.

Show More

Current UserValidated ReviewerIncentivizedSource: Seller invite

See what 533 reviewers think of Monte Carlo

4.3 out of 5 · Verified reviews from real users

[
Read all reviews
](https://www.g2.com/products/monte-carlo/reviews)