Behavioural messaging: campaigns triggered by what someone does in your product, across email, push, SMS and in-app, with segmentation over the whole event history.
Do not rebuild this at scale — but be clear about which half you need, because they are very different. **Live segments recomputed as events arrive** is a streaming problem rather than a scheduled job, and at real volume it is genuinely hard. **Multi-channel delivery** — email, push, SMS and in-app from one campaign — is four separate integrations, each with its own reliability and cost. Neither is a weekend. WHEN TO KEEP PAYING - Segments that must be current within seconds, driving time-sensitive messages - Push and SMS alongside email, at volume - Marketing people building campaigns without a developer WHEN TO BUILD - A few lifecycle sequences triggered by events from your own application, sent by email. That is most of what most products need, and the shape is covered under Loops elsewhere in this catalogue THE TWO IDEAS TO TAKE WHATEVER YOU DECIDE **One contacts table and one sends table.** Every message — campaign, sequence, transactional — is a row in the same place. That single schema decision is what lets you ask how many messages somebody has had today before sending another **A frequency cap enforced in the send path**, per person per day and per week, for anything not transactional. Transactional always goes; a receipt is never suppressed by a marketing cap. This costs an afternoon and it is the difference between a product that communicates and one people mute AND THE STOPPING RULES - A sequence is stopped by an event, not only by finishing. 'Created first project' ends onboarding, and a sequence that keeps nudging somebody who already did the thing is the fastest way to be marked as spam - Every enrolment records why it started and why it stopped - One delivery row per person per message, written before sending, so a resume is exact and a duplicate is impossible SEGMENTS WITHOUT STREAMING - Compute them at query time from the event table. At a few million events that is fast, it is always current by definition, and it cannot go stale - Materialise only the ones that are demonstrably slow, with the refresh interval visible next to the count so nobody mistakes an hourly segment for a live one THE ONE-LINE VERSION Buy it for streaming segments and four channels. Build it for email lifecycle — and take the frequency cap either way.
What you lose
- Live segments recomputed as events arrive, which is a streaming problem rather than a cron job
- Multi-channel delivery — email, push, SMS and in-app — with one set of frequency rules across all of them
- Journey branching with waits, goals and holdout groups, and the reporting to tell whether any of it worked
- Deliverability at volume on dedicated IPs, warmed and monitored
If you would rather not build
- Loops, for a smaller version of the same idea
What it costs
read from their page 15 Aug 2026
| Plan | Billed monthly | Billed yearly | Last read |
|---|---|---|---|
| — | $100/mo | — | 15 Aug 2026 |
Their pricing page is where these came from. Seeing a different price? Tell us.
The escape hatch
open source · no votes, no paid placement
Novu
$0Notification infrastructure across email, push, SMS and in-app, self-hostable.
novuhq/novufree · open source
Listmonk
$0Self-hosted campaigns with an API to trigger sends from your app.
knadh/listmonkfree · open source
Why this verdict
our own opinion · changed only by a person
32/100
Verdict no at 32. A few triggered emails are easy; live segmentation, multi-channel and dedicated-IP deliverability are infrastructure you would be rebuilding badly.
History
tracked since 10 Aug 2026 · nothing is ever overwritten
Questions about Customer.io
answered from the record above
Is Customer.io free?
No — the plan we track is $100 a month. Essentials from around $100/month billed monthly for 5,000 profiles, rising with profile count.
Can you replace Customer.io by building your own?
KEEP IT. The value is the network, the data or the infrastructure. Keep paying. Replacement score 32 out of 100, build time longer than it saves. Read what you lose before you decide.
How much does Customer.io cost?
$100 a month on Essentials — $1,200 a year. Recorded 9 Aug 2026.
What do you lose by replacing Customer.io?
Live segments recomputed as events arrive, which is a streaming problem rather than a cron job; Multi-channel delivery — email, push, SMS and in-app — with one set of frequency rules across all of them; Journey branching with waits, goals and holdout groups, and the reporting to tell whether any of it worked; Deliverability at volume on dedicated IPs, warmed and monitored. If any of those carry weight for you, keep paying.
Is there an open-source alternative to Customer.io?
Yes: Novu, Listmonk. The prompt on this page is for when you want it your way instead.
Related entries
same category first, most replaced first
Every week, something stops being worth paying for.
New verdicts, prices that moved, entries added. One email a week. Unsubscribe in one click. Nothing is being sent yet — your address is kept here, and the first issue is the first thing it is used for.
free forever · no tracking pixel · stored here, never passed to anyone

