HowlOps
FeaturesHow it worksPricingIntegrationsDocsChangelogContact
Log inStart free
PRODUCT CHANGELOG

What’s new for you.

The features and fixes that change how you use HowlOps, in plain language, release by release. For the longer stories behind each release, see What’s New.

v1.27

JUL 23, 2026

FIXED
  • Ticking "remember me" at login now keeps you signed in for up to 30 days even when your account uses two-factor authentication. Before, accounts with 2FA got a browser-session login that ended as soon as you closed the browser or when we shipped an update, so you had to sign in and re-enter a code far more often than you should have. (Safari shortens saved logins to about a week of its own accord, so there you will be asked again sooner.) A login or passkey step that is in progress, and the lockout after too many wrong codes, now also carry across our own service restarts.
  • Your daily digest and weekly uptime report emails now show real per-monitor uptime again. A backend issue meant the "monitors that struggled" and "worst monitor" sections were quietly missing and the weekly email always claimed "no check data" even when your monitors were running fine.
v1.28

JUL 24, 2026

FIXED
  • Inviting a teammate with a custom role now assigns that exact role instead of quietly falling back to the default member role.
  • Voice call notification channels can now be created without hitting an error.
  • Removing a teammate now also clears their on-call rotation slots and paging assignments, so they stop getting alerted after they have left.
  • A monitor that recovered while grouped with other simultaneous outages could go quiet, with no "back online" notification until every other affected monitor also recovered. Recovery alerts now always reach whoever was paged, the moment that specific monitor comes back up.
  • New support tickets now notify you by email as well as Telegram, so a customer's first ticket does not slip past you.
  • Batched email digests held for quiet hours now flush at the right time for your workspace's timezone instead of UTC, so they no longer arrive at the wrong hour or wait an extra day.
  • Password-protected status pages now show the scheduled-maintenance banner and let visitors subscribe once they have unlocked the page with the correct password. Both used to silently do nothing there.
  • Editing a notification channel from the Integrations page no longer silently resets its custom alert-throttle setting back to the default.
  • On-call escalation steps that page someone by email or SMS were silently failing to deliver. They now correctly reach the assigned person.
  • Exporting your data now works end to end. Request an export and download it in the app once it is ready, instead of waiting for a link that never arrived.
  • Every team member can now export their own personal data, not just workspace owners and admins.
  • An escalation step can now page multiple people and channels at once — add as many targets as a step needs, and every one of them shows up in Alert Activity and the Delivery Trace.
  • The person who created a workspace now keeps a protected, permanent line back into it. Ownership can be handed over, but the founder can no longer be quietly removed or locked out, and can always reclaim access if needed.
  • You now get an email if you are removed from a workspace or your role there changes, so account changes never happen silently behind your back.
  • Removing or downgrading a capability under Settings → Plan now clearly shows the change is scheduled for the end of your current billing period, and gives you a one-click "Keep current plan" option to undo it. Before, the confirmation could show a misleading error even though the change had actually been scheduled correctly.
v1.26

JUL 21, 2026

IMPROVED
  • When an alert opens, its start time is now backdated to the first failing check, not the moment the failure was confirmed. The incident timeline and the response-time chart mark the problem from when it actually began, so a short confirmation delay no longer makes an outage look like it started later than it did.
NEW
  • Optional slow-response warnings. If you set a slow-response threshold on a monitor, a check that comes back up but slower than that now sends a low-severity "degraded" warning rather than staying silent. It never opens an incident, never pages or escalates, respects quiet hours, and is rate-limited so a slow service does not flood you. It is fully opt-in and you can turn the degraded event off per channel in your notification preferences.
v1.25

JUL 20, 2026

NEW
  • The scheduled uptime and alerting report (a weekly or monthly email summarising uptime, incidents and average time-to-recover across your monitors) is now available on every plan, not just paid Uptime. Turn it on under Settings, Notifications, Digest.
  • Your notification preferences grid now shows your workspace's own chat channels (Slack, Mattermost, Teams and so on) instead of a fixed Slack column, so you only toggle the ones you actually use.
  • A new Cost forecast card in Settings, Billing shows your usage spend so far this period, a projected total at period end (from your current run-rate), a next-month estimate, and a six-month usage-cost trend, so you can see how your on-call usage translates into cost over time.
  • You can now acknowledge an alert by replying ACK to the alert text message. It stops the escalation right from your phone, the same one-tap acknowledgement you already had on voice calls, so whoever is on call can quiet a page without opening the app.
  • Mattermost and Microsoft Teams alerts now carry Acknowledge, Resolve and Silence buttons, matching Slack and Telegram. A click acknowledges, resolves or silences the incident straight from the channel, so your team can act on a page without leaving chat.
  • Billing now supports Czech invoicing details. Under Settings, Billing you can add your IČO (business ID, separate from your DIČ/VAT) along with a company name and billing address, and they appear on your invoices. The tax-ID list also now offers Czech DIČ directly.
FIXED
  • Maintenance windows set with a duration (rather than a fixed end time) now correctly silence alerts for their whole length, so a scheduled window no longer pages you partway through.
  • An incident's flap count no longer creeps up while a monitor is simply down for a long stretch; it now counts only genuine up-and-down flapping. Incidents you close by hand also stop escalating immediately, closing a brief window where a just-closed incident could still page the next responder.
  • Screenshot capture works again. An internal authentication mismatch was stopping captured screenshots from being saved, so manual and incident screenshots silently failed; they now go through and attach to your monitors and incidents as expected.
  • If you are on a paid plan and get a trial of a higher plan, you now return to your paid plan when the trial ends, not to Free. The "what changes when your trial ends" page reflects this too, showing the difference against the plan you actually revert to rather than always against Free.
IMPROVED
  • Each delivery receipt on the alert detail page now shows more of the story behind it: which escalation step and policy sent it, the routing rule or on-call schedule involved, and the reason and time, now visible on mobile too instead of only on wide screens.
  • The "What changes when your trial ends" page is now honest across every plan. It used to compare only a handful of limits, so some trials looked like nothing would change. It now checks your whole plan against Free and shows every difference that actually applies (status pages, check frequency, maintenance windows, retention and the rest), or a clear "nothing changes" when that is genuinely true.
  • Verifying your phone number is now done by SMS instead of an automated call. You enter your number, we text you a 6-digit code, and you type it back to confirm. The verify step now sits right where you add your phone, both on your profile and in Notifications, so a number you add is easy to confirm before your alerts rely on it. Verifying a number also keeps your profile phone in sync, so the two can no longer drift apart.
  • The alert and incident detail page is tidier. The separate "Alert activity", "Delivery trace" and "Timeline" cards are now one "Alert activity" section that tells the whole story in order, from the moment the alert arose to when it ended, together with every notification attempt and its delivery receipt. The "Check details" box that used to appear twice is now a single card.
  • A brand-new alert chain now starts pre-filled with your four personal channels (push, email, SMS and voice) in the recommended fallback order. Each row switches on only once that channel is set up, so nothing pages a place it cannot reach.
  • Settings, Roles now lists the built-in Owner, Admin and Member roles as read-only entries you can open to inspect their permissions, and clone as a starting point for your own custom role.
  • The Analytics response-time chart now shares the same polished look as the monitor detail and reports charts.
v1.24

JUL 18, 2026

NEW
  • HowlOps is now available in Czech. The whole app — marketing pages, dashboard, forms, tooltips and validation messages — switches to Czech automatically for visitors from Czechia, or manually from the language switcher in the top navigation and in Settings. English stays the default everywhere else, and you can flip back to it any time from the same switcher.
  • Frequently-asked questions now show in your chosen language.
v1.23

JUL 16, 2026

NEW
  • You can now drag across the monitor response-time chart to zoom into a specific time span. The chart, the recent-checks list, the probe map, and the incident markers all refocus on the range you selected, and a "Reset zoom" button returns to the chosen window. Zooming stays within your plan's history limit.
  • The time window on a monitor is now a single range control (Grafana-style): the button shows the current window and opens a menu with quick ranges (last 5 minutes through last 30 days) plus an absolute From/To. The presets 1h/24h/7d/30d live inside it. The whole page follows the range you choose, just like drag-to-zoom.
FIXED
  • Switching the time window on a monitor's detail page (1h / 24h / 7d / 30d) now always redraws the response-time chart for the window you picked. Previously a slower range could finish loading after a faster one and quietly snap the chart back to the old window, so it looked like switching "did nothing" or updated only after a delay.
  • A batch of small visual fixes on the monitor and analytics pages: incident markers on a zoomed-in response chart are clickable again; the mouse wheel over the chart no longer starts a zoom that never finishes; the "Reset zoom" button now matches the range picker's size and uses the regular date font; the country flag no longer spills outside its badge when you pick monitor regions; and the trend lines on the analytics KPI cards are no longer clipped at the top or bottom.
  • Times on a monitor's detail page now all use the same timezone. The response-time chart used to label its x-axis in your browser's local time while the range picker and recent-checks list used your workspace timezone, so the same moment could read as two different times. All three now use the workspace timezone, and the chart caption names the zone so it's unambiguous.
IMPROVED
  • The monitor response-time chart has a cleaner, more modern look: smooth curves, a soft gradient, and a y-axis that starts at zero and stretches to fit the highest point, so you can see where the value sits without a tall spike getting cut off. In multi-region view each region is its own clearly-colored line, and long stretches with no data now break the line instead of drawing a misleading flat or sloping segment across the gap.
  • The shaded incident markers on the response-time chart are now clickable — click one to jump straight to that incident, even a brief incident that sits between two checks (the click target is now easier to hit). The duplicate incident list that used to sit directly under the chart is gone; the "Recent incidents" panel is now the single place to browse them.
  • Live updates on a monitor's detail page now refresh only the part that changed (status, incidents) instead of reloading the whole page, and they no longer interrupt you while you're dragging to zoom the chart.
  • The probe map now opens zoomed to the regions your monitor actually checks from, instead of a whole-world view with one distant pin, and you can expand it to a taller view when you want more detail.
  • Probe regions now show a country flag: a flag with a short code (like FRA or DUB) in the monitor header, and a flag beside each region name when you create or edit a monitor. Hover the header badge to see the full region name.
  • The 90-day uptime bar now looks right for monitors younger than 90 days — it shows just the days you have instead of a mostly-empty strip — and hovering any day shows that day's date and uptime.
  • Response times now switch units to stay readable: values under a second show in milliseconds (e.g. 480 ms) and larger values in seconds (e.g. 1.50 s or 8.0 s), on both the response chart and the monitor summary. A big latency spike no longer shows as a hard-to-read four- or five-digit millisecond number.
v1.22

JUL 15, 2026

NEW
  • Your notification settings now flag a phone number that has opted out of SMS: if someone replies STOP to your text alerts, an "SMS opt-out" badge appears next to that number, so it's clear why SMS isn't being delivered there. Voice calls to the same number are unaffected.
IMPROVED
  • The monitor probe map is now a real interactive world map: pan and zoom a proper map with light or dark tiles that match your theme, in place of the previous flat illustration. Pins still color by region status (up, slow, down, or no recent data) and open that region's recent checks when clicked, and regions on higher plans still appear as locked pins you can unlock with multi-region.
v1.21

JUL 14, 2026

NEW
  • Switching from another tool? The welcome wizard now imports your existing monitors right in the first step — pick Opsgenie, UptimeRobot, Pingdom, or Better Stack, paste a read-only API key, choose what to bring over, and watch it import. Nothing in the source is changed, and re-running it later never creates duplicates.
LAST UPDATED JUL 16, 2026