# How Softr left Datadog and debugs every AWS alarm in Better Stack

[case-study-intro]
text = "Softr is the no-code platform that more than 7,000 organizations use to build client portals and internal tools. Its engineers moved their logs from Datadog to Better Stack and now use it every day to investigate production and keep the platform running smoothly."
name = "Artur Mkrtchyan"
role = "Co-founder & CTO, Softr"
avatar = "https://imagedelivery.net/xZXo0QFi-1_4Zimer-T0XQ/ea25821b-79eb-4f16-628a-ce37af282500/orig"
[/case-study-intro]

[highlights]
[[highlight]]
value = "7,000+"
label = "Organizations build their apps on Softr"

[[highlight]]
value = "Daily"
label = "Used across the engineering team, many times a day"

[[highlight]]
value = "4+ years"
label = "On Better Stack since leaving Datadog"
[/highlights]

## Where they started

Softr runs on top of the data teams already have, in Airtable, Google Sheets, PostgreSQL and similar sources, which is what its customers build their client portals and internal tools from.

Artur Mkrtchyan, Softr's co-founder and CTO, ran Softr's logging on Datadog before moving to Better Stack in 2022. Datadog is the name many teams start with, and Softr's experience was a common one: the interface was harder to work with than it needed to be, and it was expensive.

When the team looked for a replacement, Better Stack was the obvious choice, for the product experience, the platform cost and the customer support. With a shared Slack channel, the team gets help from Better Stack's engineers whenever it has a question, product feedback or needs a hand with a setup.

[quote]
text = "Many teams go to Datadog because of the brand and the name. Everyone is unhappy about the cost, but some teams don't take action until it becomes a bigger problem."
name = "Artur Mkrtchyan"
role = "Co-founder & CTO, Softr"
avatar = "https://imagedelivery.net/xZXo0QFi-1_4Zimer-T0XQ/ea25821b-79eb-4f16-628a-ce37af282500/orig"
[/quote]

## How they set it up: logs first

Softr runs its platform on AWS and sends its application logs to Better Stack, stored in Better Stack's German data region. Logs are the part of Better Stack the whole engineering team relies on. Because Better Stack is built on the OpenTelemetry standard, logs arrive in one open format whichever service sends them, so the team can search across everything in one place instead of stitching tools together.

The team works in an agent-first way. Most of Softr's engineers use Claude Code as their coding agent, and much of their tooling is reached through MCP servers rather than web interfaces.

[wide-image]
src = "https://imagedelivery.net/xZXo0QFi-1_4Zimer-T0XQ/93c6cdba-8d47-471c-750f-053b3887aa00/orig"
alt = "Production logs from Softr's platform in Better Stack's live tail"
caption = "Softr's engineers search production logs in Better Stack whenever an AWS alarm fires."
width = 2000
height = 1189
[/wide-image]

## When an AWS alarm fires

Most investigations at Softr start the same way. An alarm fires in AWS, and an engineer goes straight to Better Stack to analyze the logs and see what is going on.

Artur credits Better Stack's interface for making that loop faster than it was before. The tool is easier to configure, easier to understand and easier to use, so engineers spend their time on the problem instead of on the tool.

[quote]
text = "Having a great product with better UX, easy to configure, easy to understand and easy to use, has been a game changer."
name = "Artur Mkrtchyan"
role = "Co-founder & CTO, Softr"
avatar = "https://imagedelivery.net/xZXo0QFi-1_4Zimer-T0XQ/ea25821b-79eb-4f16-628a-ce37af282500/orig"
[/quote]

## What it costs now

Cost was one of the two reasons Softr left Datadog, and Artur has no regrets about the move. The bigger change is in how the team works: log analysis and incident response are more efficient, because the tool no longer gets in the way.

## What is next

Artur wants his tools to come to him. Today he asks the logs what needs fixing; next, he wants Better Stack to point out what should become a ticket or a pull request. Better Stack's new Claude Code plugin does that from the other side: it checks a team's setup every four hours, flags what is broken, and can open tickets and pull requests for fixes.

Dashboards and alerts are next on Softr's list. The team is also looking at error tracking, which takes Sentry's SDKs as is, and session replays, to bring more of its tooling into one place.

[quote]
size = "xl"
text = "There is no day that we as a team are not using Better Stack. We rely on it many times per day."
name = "Artur Mkrtchyan"
role = "Co-founder & CTO, Softr"
avatar = "https://imagedelivery.net/xZXo0QFi-1_4Zimer-T0XQ/ea25821b-79eb-4f16-628a-ce37af282500/orig"
[/quote]

[cta]
title = "Bring your logs into Claude Code, like Softr."
text = "Send your logs to Better Stack and query them from your coding agent through the MCP."
[/cta]
