Sign-in MFA Activity by Factor
The volume of MFA challenges across the factors the public can use to sign in, with each factor’s share of the total.
Where All MFA Activity by Factor covers every factor type the platform records, this metric keeps only the ones a member of the public can present as a second factor when signing in: sms_otp, voice_otp, and, since passkeys became generally available on 2026-07-06, fido2. The internal and administrative factors that dominate the platform-wide view are left out.
IBM Verify files MFA as a subtype of the authentication event rather than as an event type of its own, so these counts are a slice of the CanadaLogin-wide authentication events, not an independent stream.
Two factors that reach the public are deliberately excluded:
email_otpis offered only to validate an email address at sign-up; it cannot be used to sign in, so counting it here would overstate sign-in factor usage.kq,totp,qr, andibm_verify_pushare used internally or for testing, not part of the public sign-in flow.
Calculation
Restrict to the sign-in factors (sms_otp, voice_otp, fido2), then, as in All MFA Activity by Factor, de-duplicate before counting: the OTP factors (sms_otp, voice_otp) record both a sent event and a subsequent success/failure event for the same challenge, so only the sent row is counted for these; fido2 (passkeys) has no sent event and is counted through its success/failure rows. Sum count grouped by mfa_type, then divide each factor’s total by the combined total to get its share.
SELECT
mfa_type,
SUM(count) AS mfa_challenges,
CAST(SUM(count) AS DOUBLE)
/ NULLIF(SUM(SUM(count)) OVER (), 0) AS factor_share
FROM ibm_verify_events_transformed.events_mfa_activity
-- set the reporting window; see the timespans note in cookbook/notes.qmd
WHERE CAST(from_date AS DATE) BETWEEN DATE '2026-01-01' AND DATE '2026-05-28'
-- restrict to the factors the public can use to sign in. email_otp is
-- excluded because it only validates emails at sign-up, and kq, totp, qr,
-- and ibm_verify_push are internal factors, not public sign-in options.
AND mfa_type IN ('sms_otp', 'voice_otp', 'fido2')
-- de-duplicate OTP challenges: count only "sent" for the OTP factors
-- (sms_otp, voice_otp), and "success"/"failure" for fido2 (passkeys),
-- which has no "sent" event.
AND (
result = 'sent'
OR (
result IN ('success', 'failure')
AND mfa_type NOT IN ('sms_otp', 'email_otp', 'voice_otp')
)
)
GROUP BY mfa_type
ORDER BY mfa_challenges DESC;otp_types <- c("sms_otp", "email_otp", "voice_otp")
sign_in_types <- c("sms_otp", "voice_otp", "fido2")
tbl(con, in_schema("ibm_verify_events_transformed", "events_mfa_activity")) |>
# set the reporting window; see the timespans note in cookbook/notes.qmd
filter(
as.Date(from_date) >= as.Date("2026-01-01"),
as.Date(from_date) <= as.Date("2026-05-28"),
# restrict to the factors the public can use to sign in. email_otp is
# excluded because it only validates emails at sign-up, and kq, totp, qr,
# and ibm_verify_push are internal factors, not public sign-in options.
mfa_type %in% sign_in_types,
# de-duplicate OTP challenges: count only "sent" for the OTP factors
# (sms_otp, voice_otp), and "success"/"failure" for fido2 (passkeys),
# which has no "sent" event.
result == "sent" |
(result %in% c("success", "failure") & !mfa_type %in% otp_types)
) |>
group_by(mfa_type) |>
summarise(
mfa_challenges = sum(count, na.rm = TRUE)
) |>
mutate(
factor_share =
as.double(mfa_challenges) / sum(mfa_challenges, na.rm = TRUE)
) |>
arrange(desc(mfa_challenges))Interpretation notes
factor_share is a proportion between 0 and 1 and sums to 1 across these factors only, so it is not comparable to the shares in All MFA Activity by Factor, which are taken over every factor.
Do not quote factor_share as passkey uptake. It is not like for like across factors: an SMS code counts once it is sent, entered or not, while a passkey prompt counts only once it reaches an outcome, so passkeys read low. For uptake, use Sign-in Factor Share of Successes.
Passkey (fido2) activity appears only from the 2026-07-06 rollout onward, so a reporting window that predates that date will show little or none.
Alternative names
- public sign-in factor share of challenges
- sign-in-capable factor distribution