For engineering leaders
Crash and API monitoring for engineering leaders
Know how every release is doing, hear about a bad one before your customers do, and see the team recover. Release health, alerts and a Monday digest, in one place.
Plans: Release health, the weekly digest, comments and email alerts on every plan · Slack and more from Starter · tickets and Fix with AI on Growth and above · See pricing
- v3.4.198.91% crash-free users−0.61 pts vs previous · 3 new issuesRegressed
- v3.4.099.52% crash-free users+0.05 pts vs previous · 1 new issue
- v3.3.299.47% crash-free users2 new issues
Recovery time is a delivery metric
DORA measures software delivery performance with a small set of metrics. One of them is failed deployment recovery time: "the time it takes to recover from a deployment that fails and requires immediate intervention."
< 1 hour
Failed deployment recovery time for elite performers, 19% of respondents in the 2024 report.
Source: DORA, Accelerate State of DevOps Report 2024, p. 13< 1 day
For high and medium performers, 57% of respondents between them.
Source: DORA, Accelerate State of DevOps Report 2024, p. 131 week to 1 month
For low performers, 25% of respondents.
Source: DORA, Accelerate State of DevOps Report 2024, p. 13
Recovery starts when someone notices. Codeskop flags a release that regressed, alerts the right channel, and puts the evidence in front of whoever owns the fix.
What you get
- Release healthCrash-free users and sessions for each release, the new issues it introduced and the change from the previous release, on Overview. A release is flagged when crash-free users drop or a resolved issue comes back.Every plan
- Alerts before the support queueNew issues, spikes, regressions and slow or failing APIs, in Slack, WhatsApp, email or a webhook. Growth and above add Telegram, Discord and Microsoft Teams.Starter and above
- A weekly team digestEvery Monday at 08:00 in your organization's time zone: the week's top new issues by users affected, the crash-free trend, and what's waiting on each person. Slowest and most-failing endpoints from Starter.Every plan
- Know when a fix didn't holdA regression alert fires when an issue your team marked Completed happens again, and release health counts the resolved issues that returned.Starter and above
- Who's on itAssignment, status and the activity timeline on every issue, so you can see what's being fixed without asking in standup.Every plan
- The whole team, one viewQA, product, support and developers work the same issues. Unlimited team members on Growth and above, and each role gets its own starting view.Growth and above
Workflow
From a bad release to a recovered one
What happens between shipping a release and knowing it's healthy again.
- 1
Ship the release
Each release appears in release health as soon as users start on it, from the app version your SDK sends. There are no deploys to register.
Every plan - 2
See it regress
Crash-free users fall against the previous release, or a resolved issue returns, and the release is flagged as regressed.
Every plan - 3
The right person hears first
An alert rule posts the spike or regression to your channel. Assign the issue and @mention its owner, with the evidence attached.
Starter and above - 4
Watch the fix take over
The fixed release climbs in release health, and Monday's digest shows the crash-free trend for the week.
Every plan
Frequently asked questions
Does Codeskop measure DORA metrics?
No. It doesn't track deployments or calculate recovery time. It shows how each release is doing (crash-free users and sessions, new issues, regressions) and alerts you when one goes wrong, which is where recovery starts.
When is a release marked as regressed?
When its crash-free users are at least 0.5 points below the previous release, with at least 50 users on each, or when an issue your team marked Completed happens again on it.
Who gets the weekly digest?
Every member of the organization with a confirmed email, on every plan, every Monday at 08:00 in the organization's time zone. Anyone can turn it off in their notification settings or with the one-click unsubscribe link, and nobody gets an email when there's nothing to report.
What's in the digest on each plan?
On every plan: the top new issues of the week, crash-free users against the previous week, and the comments, @mentions and assigned issues waiting on you. Starter and above add the slowest and most-failing endpoints.
Do crash-free numbers need signed-in users?
No. A person is the user ID you pass to identify, or the install ID when there isn't one, so apps with anonymous users get the same numbers.
Is this APM or uptime monitoring?
No. Codeskop monitors crashes, errors and API calls from your apps and services. It doesn't trace requests across services or check whether your site is up.
Know what broke. Fix it faster.
Bring your whole team into the same issues.
Set up in minutes. Free plan for up to 5 team members, no credit card required.