USSD Funnels Guide
All suggested CleverTap funnels for the Uganda USSD service (*284*300#). Each funnel includes the exact steps, filters, and the product question it answers.
February 2026
Uganda
8 Events
11 Funnels
How to Read This Guide
Each funnel card shows:
Must Have
Essential for launch — set up in CleverTap immediately
Should Have
Important for optimization — set up within first week
Nice to Have
Deeper insights — set up once baseline data exists
Step colors: cherry = entry point, aqua = user action, mauve = decision point, lime = goal/success.
All funnels use a 30-minute conversion window (USSD sessions are short-lived).
Core Repayment Funnels
1. Overall Repayment Funnel
How many users who start a session with an active loan end up making a successful payment?
Session Start
BE_USSD_SESSION_START
status = loan_found
→
Balance Viewed
BE_USSD_BALANCE_VIEWED
→
Repayment Confirmed
BE_USSD_REPAYMENT_CONFIRMED
→
Repayment Success
BE_USSD_REPAYMENT_RESULT
status = success
Key question: What is our overall USSD-to-repayment conversion rate? Where is the biggest drop-off?
2. Full Repayment Funnel
Conversion for users who choose to pay their entire outstanding balance. This is the shortest path — no amount entry step.
Session Start
BE_USSD_SESSION_START
status = loan_found
→
Balance Viewed
BE_USSD_BALANCE_VIEWED
→
Full Confirmed
BE_USSD_REPAYMENT_CONFIRMED
payment_type = full
→
Success
BE_USSD_REPAYMENT_RESULT
status = success
Key question: What % of users choose full repayment, and what is the success rate? (Expect higher conversion than partial — fewer steps.)
3. Partial Repayment Funnel
Conversion for users who enter a custom amount. Has an extra step (amount entry) which is a friction point — compare drop-off vs full repayment.
Session Start
BE_USSD_SESSION_START
status = loan_found
→
Balance Viewed
BE_USSD_BALANCE_VIEWED
→
Amount Entered
BE_USSD_AMOUNT_ENTERED
valid = true
→
Partial Confirmed
BE_USSD_REPAYMENT_CONFIRMED
payment_type = partial
→
Success
BE_USSD_REPAYMENT_RESULT
status = success
Key question: How much does the extra "enter amount" step hurt conversion vs full repayment? Is the amount entry UX causing drop-off?
Repayment Drop-off & Failure Analysis
4. Cancellation Funnel
Track where users bail out. Users who see the confirmation screen but choose to cancel instead of confirming.
Balance Viewed
BE_USSD_BALANCE_VIEWED
→
Repayment Confirmed
BE_USSD_REPAYMENT_CONFIRMED
vs
Cancelled
BE_USSD_REPAYMENT_CANCELLED
Key question: What % of users reach the confirmation screen but cancel? Are amounts too high? Is the confirmation wording unclear?
5. Repayment Failure Analysis
Of users who confirm repayment, how many fail? Group by error_code to find the most common failure reasons.
Repayment Confirmed
BE_USSD_REPAYMENT_CONFIRMED
→
Success
BE_USSD_REPAYMENT_RESULT
status = success
vs
Failed
BE_USSD_REPAYMENT_RESULT
status = fail
Key question: What is our payment success rate? Which error codes are most frequent? Are failures concentrated on certain users or times of day?
6. Invalid Amount Entry Rate
How often do users type invalid amounts (non-numeric, too low, exceeding balance)? Indicates UX friction in the partial repayment flow.
Balance Viewed
BE_USSD_BALANCE_VIEWED
→
Invalid Amount
BE_USSD_AMOUNT_ENTERED
valid = false
then
Valid Amount
BE_USSD_AMOUNT_ENTERED
valid = true
Key question: What % of amount entries are invalid? Do users recover (retry with valid amount) or abandon? Should we show clearer validation hints?
Registration & Onboarding Funnels
7. Registration → Repayment Funnel
Users who start without a payment method: do they successfully register their MoMo wallet AND complete a repayment in the same session?
Session Start
BE_USSD_SESSION_START
status = no_payment_method
→
Registration OK
BE_USSD_REGISTRATION
status = success
→
Balance Viewed
BE_USSD_BALANCE_VIEWED
→
Repayment Success
BE_USSD_REPAYMENT_RESULT
status = success
Key question: Is inline registration effective? Or do users drop off after registering (too many steps in one session)?
8. Registration Success vs Failure
Of users prompted to register their MoMo number, how many succeed vs fail? Failures indicate the number validation or verification process has issues.
Session Start
BE_USSD_SESSION_START
status = no_payment_method
→
Registration Success
BE_USSD_REGISTRATION
status = success
vs
Registration Failed
BE_USSD_REGISTRATION
status = fail
Key question: What is the registration success rate? If low, is it a UX issue (users entering wrong format) or a backend issue (verification failures)?
Unregistered User Acquisition
9. Unregistered User Engagement
When someone who isn't a Fido user dials the USSD code, they see a menu with 3 options. Which option do they pick? Do they engage at all?
Session Start
BE_USSD_SESSION_START
status = unregistered
→
Menu Selection
BE_USSD_UNREGISTERED_MENU
Group by the choice property to see the split:
📲
Choice 1
Download the app
📞
Choice 2
Contact support
🔔
Choice 3
Request callback
Key question: Which CTA resonates most? Is USSD driving app downloads? Should we change the menu order or wording?
10. USSD-to-App Conversion (Cross-Platform)
Of unregistered users who chose "Download app" and received the SMS link, how many actually signed up? Requires joining USSD events with app signup events by phone number.
Unregistered Menu
BE_USSD_UNREGISTERED_MENU
choice = 1 (download)
→
SMS Sent
(SMS delivery log)
→
App Signup
App signup event
same phone number
Key question: Is USSD an effective acquisition channel? What is the SMS-to-signup conversion? (Requires cross-platform identity matching by phone.)
Overall Channel Health
11. Session Outcome Distribution
Not a traditional funnel — a segmentation of ALL session starts by their status. Gives a bird's-eye view of who is using the USSD channel and what they encounter.
Create as a pie chart / bar chart in CleverTap, not a funnel. Event: BE_USSD_SESSION_START, grouped by status.
loan_found
Happy path
unregistered
Not a user
no_payment_method
Needs registration
no_active_loan
No outstanding loan
blocked
Account blocked
kyc_not_verified
KYC incomplete
error
Technical failure
Key question: What % of USSD traffic is actually repayment-eligible? How much is unregistered users discovering the code? Are error rates acceptable?
Summary — All 11 Funnels at a Glance
| # | Funnel | Steps | Priority | Answers |
|---|---|---|---|---|
| 1 | Overall Repayment | Session → Balance → Confirm → Success | Must | End-to-end conversion rate |
| 2 | Full Repayment | Session → Balance → Confirm (full) → Success | Must | Full repayment adoption & success |
| 3 | Partial Repayment | Session → Balance → Amount → Confirm (partial) → Success | Must | Amount entry friction & conversion |
| 4 | Cancellation | Balance → Confirmed vs Cancelled | Should | Why users bail at confirmation |
| 5 | Repayment Failure | Confirmed → Success vs Fail (by error_code) | Should | Payment success rate & error patterns |
| 6 | Invalid Amount Entry | Balance → Invalid → Valid recovery | Should | Amount input UX quality |
| 7 | Registration → Repayment | Session (no PMT) → Register → Balance → Success | Must | Inline registration effectiveness |
| 8 | Registration Success/Fail | Session (no PMT) → Success vs Fail | Should | Registration pass rate |
| 9 | Unregistered Engagement | Session (unregistered) → Menu choice | Must | Which CTA works for non-users |
| 10 | USSD-to-App Conversion | Menu (download) → SMS → App signup | Nice | USSD as acquisition channel |
| 11 | Session Outcome Distribution | All sessions grouped by status (chart) | Must | Channel health & traffic mix |
CleverTap Setup Notes
1
Conversion Window
Use 30 minutes for all USSD funnels. Sessions timeout after 5 minutes, but allow buffer for retry flows.
2
Identity Resolution
Events use user_id when known, phone_number when not (unregistered users). CleverTap should resolve both to the same profile once the user signs up.
3
Event Naming
All events use the
BE_USSD_ prefix. They appear under this prefix in CleverTap's event list.4
Pipeline
ussd-service → SNS topic → events-service → CleverTap. Events arrive in near real-time (seconds).
5
Full vs Partial Comparison
Set up funnels 2 and 3 side by side to compare conversion rates. The key delta is the amount entry step — if partial drops off significantly more, consider UX improvements (e.g., suggested amounts).