LIVE RISK SCORING

Catch churn before it happens.

Upload a customer CSV and get a risk score for every account in seconds. No integration, no account, nothing stored after you close the tab.

accepts .csv · scores in-browser · avg. scan time 4.2s
live_scan.feed
● scanning
customer_idriskscoretier
cus_8G2LqP
0.82HIGH
cus_4Xk9Wa
0.61MED
cus_M0eZ7t
0.18LOW
cus_R2vHnB
0.74HIGH
cus_QcT10s
0.29LOW
1,204 accounts scannedupdated just now
HOW IT WORKS
01

Upload

Drop a CSV of customers. No API keys, no schema mapping.

02

Score

Every row gets a churn probability and a low / medium / high tier.

03

Act

Export the high-risk list and hand it to your CS or lifecycle team.

WHAT YOU GET
customer_idplanlast_login_daysrisk_scoretier
cus_8G2LqPpro420.82HIGH
cus_M0eZ7tfree30.18LOW
cus_4Xk9Waenterprise210.61MED
cus_R2vHnBpro580.74HIGH
cus_QcT10sfree80.29LOW
cus_wF1kBzpro140.44MED
tier_breakdown
  • LOW58%
  • MED27%
  • HIGH15%
DATA & PRIVACY

Session-only processing

Your CSV is parsed in memory. Close the tab and it's gone.

No account needed

No signup, no email, no login. Open, upload, score.

No third-party sharing

Nothing is sent to analytics, ads, or trackers.

CSV FORMAT
  • user_id
    Unique customer identifier
    required
  • last_active_days
    Days since last activity
    required
  • total_sessions
    Sessions in the last window
    required
  • plan_type
    free · pro · enterprise
    required
  • feature_adoption_score
    0–10 adoption score
    optional
note

Column names are matched case-insensitively. Extra columns are preserved in the exported file. Not sure what to send? Start from the sample.

FAQ
What counts as a valid CSV?

A plain .csv file with a header row and at least the four required columns: customer_id, tenure_days, monthly_spend, and last_login_days. Extra columns are fine — we just ignore anything we don't recognize.

Is my data used to train anything?

No. Your file is scored for this session only and isn't used to train or improve any model.

What happens if my columns don't match?

If a required column is missing, we'll tell you which one before scoring anything. If a column is named differently than expected (e.g. "last_login" instead of "last_login_days"), we match common variations automatically — otherwise you can rename the header and re-upload.

Can I re-run a scan on the same file?

Yes. Uploading again — the same file or an updated one — runs a fresh scan every time. Nothing carries over between scans.

Do you store any of my data?

No. Everything is processed for the length of your session and is gone once you close the tab.

Ready to see who's at risk?