What is a good churn rate?

It depends on who you sell to. Products sold to large companies lose far fewer customers a month than tools sold to small teams, so compare yourself with products that sell to the same kind of customer, and watch your own trend month over month.

What does 5% churn mean?

A 5% churn rate means you lose 5 of every 100 customers in that period. At 5% a month, you’d lose close to half your customers over a year unless new ones replace them.

How do you reduce churn?

Spot risk early from product behavior, act before the cancel screen, and match the save to each customer’s value. Intempt scores churn risk, starts the save journey on its own and shows what held.

How does churn prediction work?

A churn prediction model learns from customers who did and didn’t leave, then scores everyone on how likely they are to churn. With Intempt, you pick the churn event and the model learns it.

Do you ship a churn score out of the box?

No, on purpose. You pick the event that means churn in your product, and a model learns to predict that. A canned score has to guess, and churn means something different in every product.

Do I need a customer success team to run it?

No. A CS tool hands a person a list to work through. Here the score moves someone into an audience and the audience starts the journey, so it runs at any list size.

How is this different from product analytics?

Analytics tells you churn went up. Intempt scores who is at risk and acts on it in the same place, so you never export a cohort into another tool to do anything about it.

Churn reduction for growth teams

Reduce churn before the cancel click.

Q: What is customer churn?

Customer churn is when customers stop buying from you or cancel their subscription. It’s measured as a churn rate, the share of customers you lose over a set period, like a month.

Pick the event that means churn for you, score every customer on live behavior, and the save journey starts on its own.

Start for free

Behavior scores
Save journeys
Exit rules

predict('Subscription Cancelled')Data Analyst agent· scoring live

Lina Ortizacct_2817 · Kestrel StudioAt risk

Behavior · last 7 days

Logins: ↓ 68%
Boards viewed: 3 vs 21
Teammates active: 2 of 9

Churn risk74%

60%

Crossed your 60% threshold2s ago

Save journey

Need a hand?

Step 1 · email sent

Exit rule

risk < 40%

Armed

7+
years profitable,and counting
rating on G2: 4.5
events processed: 10B+
user profiles: 500M+
enterprise customers: 50+

StockInvest
“Acquisition and retention finally feel like one connected motion instead of two separate problems.”
Jim StrombergCEO, StockInvest
FieldsUSA
“We finally have one system that sees the whole customer journey.”
Eric GardnerCOO, FieldsUSA
Hoperfy
“With Intempt, we built a signal-led pipeline driven by real behaviors.”
Tadas KertenisCo-founder, Hoperfy

Your save offer comes too late to work.

Today most teams find out at the cancel screen, where the only lever left is a discount that trims the margin you saved.

Start for free

Scoring churn risk from last week’s product use

Reading product use from the last 7 days…

Data Analyst agent

Scoring churn risk on every event.

Customers scored12,480
Risk rose today96
Crossed 60% today18

Risk rising+52 pts in 3 weeks

Lina OrtizKestrel Studio

save at 60%

3 weeks ago · 22%today · 74%

Save journey starts at 60%

Churn riskLive

All accounts

Window 7 daysLoginsBoards viewedTasks createdTeammates activeTimeline view

Lina OrtizLogins ↓ 68% · 3 boards viewedMediumHigh74%
Arjun PatelTimeline view · never openedMedium63%
Sofia RossiSeats removed · 3 this weekMedium58%
Jonah ReyesLogins steady · 14 boards viewedLow31%
Maya ChenLogins ↑ 12% · 9 of 9 teammates activeLow6%
Priya NairScoring

Three levers

Three ways to reduce customer churn.

Save at-risk customers before they cancel, lead non-adopters to what they’re missing, and focus on the ones worth most.

01 · At-risk

Save customers while they’re still drifting

The churn score becomes a filter, so a save journey starts the moment someone crosses your threshold, matched to the risk.

Pick the event that counts as churn for you.
Messages scale with how high the risk is.
The save starts when they enter the segment.
Scores update from live product behavior.

intempt.com/attributes/churn-risk

Churn riskAI attributePredicts Subscription Cancelled · window 7 daysData Analyst agent

Churn event

Subscription Cancelled

Behavior window

7 days

Score

Paying customers

Risk tiers · 12,480 customers scoredMoves with live product use

High≥ 70%1,1409.1%
Medium40-69%2,31018.5%
Low< 40%9,03072.4%

Message by riskLow risk gets no save message

HighPersonal email from CSMCall offerSave · high risk
MediumTips emailIn-app guideSave · medium risk

Save journey starts at60%1,486 in segmentLive

What changes

Six things you stop doing at the cancel page.

Data Analyst agentModel trained
- Churn eventSubscription Cancelled
- Signals learned46
- Refreshevery event
0data scientists staffedNot needed
Building a churn model
You pick the event and a model learns to predict it, so no data scientist has to build or maintain a churn model.
23days earlierthan the first support ticket
Risk flaggedFirst ticket
Day 0 · crossed 60%Day 23
Waiting for support tickets
Scores read product behavior as it happens, so you spot a drifting customer long before they ever open a ticket.
Churn risk ≥ 60%
At-riskAudience · +18 today1,486
Save journeyruns on entryStarted
Handing lists to a CS team
A score moves someone into an audience and the audience starts the journey, so it runs at whatever size your list is.
Weeks before
- Tips email4 wk
- Feature tour3 wk
- CSM call2 wk
- Right-size plan1 wk
Cancel screen
Cancel plan?
Before you go…
20% offfor 3 months
Cancel anyway
Discounting at the last minute
Acting while customers drift gives you more levers than the cancel screen, where a discount is all that’s left.
0CSV exportUpload to email tool
ScoreSaveSame place
Exporting cohorts to act on
Scoring and saving happen in one place, so you never export an at-risk cohort into another tool to act on it.
CS platformper account$10,700/yr
One profileLina OrtizIncluded
Churn scores74%
SegmentsAt-risk
Save journeysStep 1
Paying for a CS platform
Scores, segments and save journeys share one profile, so you skip a customer success platform priced per account.

The difference

The save starts
weeks before the cancel.

Before and After
No.	Before	After
01	You find out a customer is leaving when they click the cancel button.	A churn score flags them weeks earlier, from how they use your product.
02	Your CS tool hands someone a list, and it waits in their inbox for days.	The score moves them into a segment, and the save journey starts itself.
03	A generic churn score guesses what churn means for your business.	You pick the event that means churn, and the model learns that one.
04	You can’t tell which save play actually kept your customers around.	Retention by week shows which save plays kept your customers around.
05	Everyone at risk gets the same discount, whatever they’re worth to you.	The save list is ordered by value, so top accounts get attention first.

The offer

What saving customers costs the old way.

This is what a typical team pays when the CS platform, the churn model and the tool that sends each save are bought apart.

The old way

$17,900/yr

Year one, three CS seats and 5,000 profiles.

Intempt

$49

per seat a month

Email platform³$100/mo
Custom churn model²$6,000
CS platform¹$10,700/yr

The old way

$49 · per seat a month

Intempt

Built in

Save journeys

The retention stack you were about to buy

A churn score you define.
Save journeys that run alone.
Save lists ordered by value.
Retention by day, week, month.

Runs on its own

The CS manager you were about to hire

Lifecycle Marketer agent
Runs each save journey.
Data Analyst agent
Keeps churn scores fresh.
AI Attributes
Stores each churn score.
Retention
Shows who came back.

1Vendr data via ChurnTools and Gaintrace (2026). ChurnZero Professional for three users has a median of about $10,700 a year.
2Fast Data Science, Reforge (2026). Custom churn prediction models typically take 20-40 hrs of data science consulting at $100-250/hr ($2,000-10,000); this uses $6,000.
3Customer.io pricing (eesel AI, Encharge, 2026). The Essentials plan is $100/mo for up to 5,000 profiles.

Control

You decide what churn means here.

You choose the churn event, the risk threshold and the exit rules, and every save journey runs inside the rules you set.

churn event: Subscription Cancelled
window: 7 days
start at: 60% risk
exit below: 40% risk

Your event, not a canned one
No churn score ships out of the box, so the model learns what churn means in your product, not someone else’s.
Thresholds you set
Choose the risk level that starts a save, and match each message to how high that risk has climbed for them.
Exit rules for every journey
Set what pulls someone out of a running journey, so a save never lands on top of an upsell they’re getting.
Deliverability protected
Suppression keeps a big save campaign from hurting your sender score, so every other campaign still lands.

Integrations

Wired into the tools your team already uses.

See all integrations

JavaScript
Apache Kafka
Slack
Stripe
Amazon S3
Twilio
Gmail
iOS
SendGrid
Webhooks
Node.js
Google Calendar
JavaScript
Apache Kafka
Slack
Stripe
Amazon S3
Twilio
Gmail
iOS
SendGrid
Webhooks
Node.js
Google Calendar

Go deeper

Two more ways to keep customers longer.

One page grows the value of every customer tier, and the other brings back signups who stalled before getting value.

Maya ChenMeridian · 14th order · just nowTier up
Predicted LTV$2,840Value tierLoyalChampionsorder #14
20% off · not sent
VIP early accessJourney started
VIP homepageLive for Champions
Job 01
Customer LTV
Score every customer by value and give each tier its own journey, offer and site experience, without blanket discounts.
See how LTV tiers work
Tom BeckerBluefin Labs · Signed up 6 days agoStalled
Setup progress1 of 4
Create boardInvite teamAdd tasksSync calendar
No progress in 5 days
Activation journeyTom Becker · Bluefin LabsStarted 2s ago
Finish your setupSMSPush
Home screen changes on returnSetup checklist up top
Job 02
User Activation
Find everyone who stalled before activating, reach them by email, SMS or push, and change what they see when they return.
See how activation works

The products underneath

Each one reads the same customer profile

Frequently asked questions

Straight answers about reducing churn, churn rates, churn prediction and how Intempt starts every save on its own.

Customer churn is when customers stop buying from you or cancel their subscription. It’s measured as a churn rate, the share of customers you lose over a set period, like a month.

Ready when you are

Know who’s leavingwhile you can act.

Connect your product and billing data, and churn scores and at-risk segments go live in the same session you set them up.

Connect your AI to your customer data

$ claude mcp add intempt -- npx -y @intempt-technologies/mcp

Reduce churn before the cancel click.

Your save offer comes too late to work.

Three ways to reduce customer churn.

Save customers while they’re still drifting

Six things you stop doing at the cancel page.

Building a churn model

Waiting for support tickets

Handing lists to a CS team

Discounting at the last minute

Exporting cohorts to act on

Paying for a CS platform

The save starts weeks before the cancel.