KRAKEN

Pass Rate Performance
Deep Dive by Country

Data Source Analysis & Orchestration Recommendations

Prepared by FrankieOne March 2026 CONFIDENTIAL

Executive Summary

53%

Overall STP (Q1 2026)

58

Countries at 0% STP

32 + 57

Pre-Existing + New Markets

17%

Socure API Error Rate

The −4pp drop from Q4’s 57% is entirely new-country dilution, not core market regression. UK 78.6%, Germany 79.3%, Denmark 72.8%, Sweden 70.7% - established markets remain healthy or improving. The 53% overall figure reflects ~1,000/day failing entities in zero-pass new markets.

Key Findings (Updated 8 Mar 2026)

  • ●Overall STP pass rate Q1 2026: 53% (Q4 2025: 57%). The −4pp reflects new country expansion diluting the overall rate. Core markets all ≥70%.
  • ●Only 5 of 30+ newly enabled countries have achieved any meaningful pass rate. Finland (39%) and Portugal (20%) are standout new cohort performers.
  • ●~1,000 Kraken end-users per day are attempting verification in zero-pass markets and receiving hard failures. This is a real UX and compliance risk.
  • ●58 countries at 0% STP. The causes fall into three root themes: poor input address quality, missing National ID submission, and datasource coverage gaps.

Top 3 Strategic Priorities

1
Address capture uplift - house number missing on 37% of GDC requests. Single highest-impact fixable lever. Owner: Kraken Eng + Product
2
National ID onboarding uplift - 9 new markets at guaranteed 0% FM because NatID is mandatory and not being submitted. Owner: Kraken Eng + Product
3
Reroute coverage gap markets - GDC & Socure. Where GDC has no viable coverage, FrankieOne will reroute. Owner: FrankieOne (routing + vendor)
→ View full country detail (Part B)

Quarterly Performance Trajectory

The path from 53% back to 60%+ is clearly mapped. The problems are diagnosed and the fixes are defined.

53% Overall STP Q1 2026 ▼ 6.45% vs Q4 2025 (57%)

Driven by new country expansion diluting the overall rate, not by deterioration of core markets.

Quarter Pass Rate STP Context vs Prior Qtr
Q2 2025 & prior ~48% (low) FrankieOne tightened ruleset - Kraken required to match on full address (Street + House Number + City + State OR Zip). Previously passing on 2 elements only (City + Zip). Temporary dip to ~48% as stricter standard applied. Baseline reset
Q3–Q4 2025 48% → 57% 10pp improvement over two quarters as FrankieOne optimised routing, datasource coverage, and NatID submission for established markets. Q4 2025 peaked at 57% - strongest quarter to that point. ▲ +10pp
Q1 2026 53% Expansion event ~20–24 Feb added 57 new markets (135 total active, 32 pre-existing). New country failures are diluting the overall rate - core markets are holding. ▼ ~4pp
Q2 2026 (forecast) 60%+ target If house number fix deployed (single Kraken engineering change), estimated +8–12pp uplift across GDC markets alone. Combined with NatID submission for 9 critical new markets and deduplication of wasted API calls, 60%+ overall STP is achievable. ▲ +7pp est

Source: ThoughtSpot Quarterly Pass Rate STP liveboard - Account: Kraken, Entity Type: Individual, Last 12 months, Timezone: Australia.

Three Root Causes - Optimisation Framework

These three issues explain the vast majority of 0% pass rates across new markets. They are data submission and routing configuration gaps that are fully addressable. The analysis below covers new country markets specifically.

ISSUE 01

Poor Input Address Quality

House number is absent on 37% of all GDC requests globally - the single highest-impact fixable lever across every GDC market simultaneously.

Italy: −54pp FM loss without house #

Germany: −53pp FM loss

Poland: −35pp FM loss

Fix: Kraken Eng to populate houseNumber field in API payload. Single change, all GDC markets benefit.

ISSUE 02

National ID Not Submitted

9 new markets at guaranteed 0% FM because NatID is mandatory and not being submitted. This is a Kraken onboarding/data collection gap.

India: 4 source-specific ID types needed

South Africa: SA ID 13 digits mandatory

Colombia: Cédula 10 digits not sent

Fix: FrankieOne to provide NatID requirements guide by country (ID type, format, field name, examples). Kraken to update onboarding UX.

ISSUE 03

Datasource Coverage Gaps

Where GDC has no viable coverage (KE, VN, AE, HK, MY, SG), FrankieOne will reroute. Where Socure has no coverage, same approach.

6 markets: no GDC address source at all

14 markets: Socure account config errors

4 markets: suspected inactive datasources

Fix: FrankieOne to reroute & resolve. No Kraken engineering action required for routing changes.

New vs Pre-Existing Country Split

Understanding the distinction is critical: pre-existing countries need optimisation (address quality, vendor cascade), while new countries need enablement (NatID submission, routing setup).

32 PRE-EXISTING 89% of total volume
Core markets (UK, DE, DK, SE)≥70% STP
Mid-performers (FR, NL, ES, IT)50–67% STP
Structural zero (IN, ZA, ID, CO)0% STP

These markets are holding or improving. The 4pp Q1 drop is NOT from regression here.

57 NEW EXPANSION 11% of volume, 100% of the problem
Top performers (FI, PT)20–39% STP
Near-zero (GR, RO, TR)0.2–7% STP
Zero pass rate0% × ~50 countries

~1,000 users/day hitting hard failures. Address quality + NatID submission = primary fix levers.

Recommendation: Use the new/pre-existing distinction for the next 1–2 quarters while the expansion cohort stabilises. Different fix strategies apply to each cohort. View full country detail →

Global Pass Rate Overview

Entity-level STP pass rate by country. Hover for details.

Source: ThoughtSpot Kraken Prod Standard Liveboard v1 | Last 1 Month (19 March 2026)

Europe: The 0% Problem

Core European markets show 0% pass rates on both National ID and Tax ID data sources. Hover for country-level detail.

Country Performance Tiers

TOP (>75%)

2

countries

74,523 entities | 79% avg

Germany (81%)
UK (77%)

MEDIUM (40-75%)

15

countries

150,587 entities | 56% avg

Sweden (73%), NLD (57%), Spain (52%), Austria (50%), NZ (46%), Poland (45%), Switzerland (43%)

LOW (1-39%)

18

countries

53,309 entities | 20% avg

Finland (35%), Portugal (23%), Canada (16%), Greece (5%), Croatia (5%), Mexico (1%), Ukraine (0%)

ZERO (0%)

69

countries

46,714 entities | 0% avg

South Africa (2,941), Kenya (1,835), Colombia (1,560), Taiwan (1,258), Czech Rep (1,087) + 64 more

Pass Rate by High-Volume Country

European Data Source Failure: 0% at Check Level

At the individual check level, National ID and Tax ID return 0% across core European markets:

Country National ID
Pass Rate		 National ID
Volume		 Tax ID
Pass Rate		 Tax ID
Volume		 Combined
Volume
Belgium0.0%32,2750.0%7,28739,562
Germany0.0%9,7940.0%60,14969,943
Netherlands0.0%29,4510.0%21,83951,290
Norway0.0%4,3260.0%9985,324
Austria0.0%3380.0%1,4051,743
France0.0%2,342--2,342
UK0.5%1920.0%9881,180

170,000+ entities across core European markets are achieving 0% pass rate on both National ID and Tax ID checks. These data sources are not returning usable results for European identity verification.

Key Finding: Platform-Wide Gap, Not Kraken-Specific

We benchmarked Kraken against all other FrankieOne customers using the same data sources in the same countries. The result: every customer gets 0% - confirming this is a provider coverage gap, not a Kraken configuration issue.

Country Data Source Kraken Rate Kraken Volume Industry Avg Other Customers Verdict
BelgiumNational ID0.0%32,2750.0%3Platform gap
GermanyNational ID0.0%9,7940.0%4Platform gap
NetherlandsNational ID0.0%29,4510.0%3Platform gap
NorwayNational ID0.0%4,3260.0%1Platform gap
FranceNational ID0.0%2,3420.0%3Platform gap
SpainNational ID16.0%318,8121.0%3Kraken leads
ItalyNational ID20.9%173,1060.0%3Kraken leads

Data Source Gap Analysis

Kraken uses only 3 of 10+ available data sources on the FrankieOne platform.

Currently Configured (3)

National ID 0% in 8 EU countries
Tax ID 0% in 7 EU countries
Mobile Phone Only USA & Brazil

Available but NOT Configured

Drivers Licence
68% avg5.8M checks
Passport
66% avg1.5M checks
Birth Certificate
56% avg112K checks

Opportunity: Drivers Licence + Passport in Kraken's 0% Countries

Country Current Rate Kraken Volume Drivers Licence
(other customers)		 Passport
(other customers)		 Potential Lift
France0%2,34278.9%18.7%0% → 79%
Netherlands0%51,29070.7%17.6%0% → 71%
UK0.5%1,18070.1%32.2%0% → 70%
Switzerland0%9167.9%26.5%0% → 68%
Germany0%69,94359.6%15.6%0% → 60%
Belgium0%39,562-20.0%0% → 20%
Norway0%5,324-18.1%0% → 18%

Provider / Vendor Analysis

Kraken's checks are routed through multiple identity verification providers. Different providers have different data lakes, coverage, and success rates per country.

Provider Volume
(Daily)		 KYC Pass
Rate		 Trend Primary Markets Opportunity
worldview_one_plus
TransUnion WorldView		 49.8K 34-35% Stable EU-wide (DE, NL, ES, IT, PL, SE, BE, NO, AT, FR, DK) Evaluate alternatives for 0% EU countries
datazoo
Datazoo		 4.5K 42-47% ▲ Improving Multi-country Route more traffic through datazoo where it outperforms
Socure
Socure Inc.		 3.8K 17-22% ▲ Improving USA, Poland Low performer - use as fallback only, not primary
Equifax CA
Equifax Canada		 1.8K 5-7% Stable (low) Canada only Primary reason for Canada's 11% - add/swap provider
centrix
Centrix NZ		 180 26-36% ▼ Declining New Zealand Small volume - monitor or add backup
GDC
Global Data Consortium		 - 31% - Multi-country Evaluate coverage in EU zero-pass countries

Key Vendor Insights

  • ●Datazoo outperforms WorldView by 8-12 percentage points (47% vs 35%). Routing more traffic through Datazoo where coverage overlaps could improve overall rates.
  • ●Equifax CA is the root cause of Canada's 11% pass rate. At 5-7%, this provider is effectively non-functional. A replacement or supplementary Canadian provider is critical.
  • ●WorldView handles 80%+ of all checks but at only 34%. A multi-vendor cascade (try Datazoo first where available, fall back to WorldView) could lift the overall rate by 5-10%.
  • ●Socure at 17-22% is improving but still underperforming. Monitor the trend - if it reaches 30%+ it becomes a viable secondary provider.

Immediate: Fix Canada

Replace or supplement Equifax CA (5-7%) with a higher-performing Canadian provider. Other FrankieOne customers achieve 73% on Drivers Licence and 33% on Passport in Canada.

Quick Win: Rebalance Traffic

Where Datazoo and WorldView both have coverage, route primary traffic through Datazoo (47%) instead of WorldView (35%). Use WorldView as fallback.

Strategic: Multi-Vendor Cascade

Implement Datazoo → WorldView → GDC cascade for EU. For each country, determine which provider has the best local data lake and route accordingly.

Root Cause Deep Dive - Address Quality Impact

Measured from live GDC API logs (6,229 transactions, 6–8 March 2026). House number is the single highest-impact fixable lever.

1. House Number Missing - 37% of GDC Requests Globally

Kraken’s pass standard requires: Street + House Number + City + State OR Zip/Postal Code. House number is absent on 37% of all GDC requests - when present, FM rates jump by 21–54 percentage points. This is a single Kraken engineering change that benefits every GDC market.

Italy −54pp, Germany −53pp, Poland −35pp, Sweden −31pp FM loss without house #

2. National ID Not Submitted in Mandatory Markets

40 of 57 new expansion countries require a National ID that Kraken’s API is NOT submitting. This is the single biggest structural lever for new markets - it’s not a FrankieOne problem, it’s a Kraken onboarding/data collection gap. Israel achieves 73% FM on non-error transactions when NatID is present.

9 markets at guaranteed 0% FM - NatID mandatory but 0% submitted

3. Vendor Coverage Gaps & Cascade Routing

Where GDC has no viable coverage (KE, VN, AE, HK, MY, SG), FrankieOne will identify and switch to the best available vendor. 14 countries blocked by Socure account configuration errors. 4 GDC markets (Bulgaria, Hungary, Lithuania, Turkey) submitting data correctly but returning 0% FM - suspected inactive datasources.

No Kraken engineering action required for routing changes

4. Socure API Errors - 17% Hard Failure Rate

17% of all Socure API calls are returning hard HTTP errors (2,405 matched transactions analysed). FrankieOne to work with Socure to fix validation issues. Includes ZIP format errors (Bolivia, Angola causing 100% API failures) and unactivated country accounts.

FrankieOne to resolve with Socure AM

Recommendations

Tier 1 - Quick Wins (0-30 Days)

Enable Drivers Licence

Platform data proves DL achieves 60-79% in FR, NLD, GBR, CHE, DEU. Single biggest impact change.

0% → 60-79%

5 European countries

Enable Passport Fallback

Passport achieves 15-32% across EU markets. As secondary source after DL, provides incremental lift.

+15-32%

fallback coverage

Cascade Orchestration

Configure: National ID → Drivers Licence → Passport → Manual Review multi-source flow.

+20-40%

incremental lift

Tier 2 - Medium Term (30-90 Days)

Optimise US SSN Stack

1.2M entities at 55%. A 10% lift = 120K fewer manual reviews. Add secondary SSN validation sources.

55% → 65-70%

Regional Profiles

Replace one-size-fits-all with EU (DL+Passport+eID), Americas (SSN+Mobile), APAC (Registry) profiles.

37% → 55-60%

Mobile Phone in EU

Currently only USA + Brazil. Carrier-based verification can supplement document sources in European markets.

+5-15%

Projected Impact

Tier 1 alone (adding Drivers Licence + Passport) is projected to lift overall STP from 37% to 50-55% - a 35-50% reduction in manual review volume. Tier 2 optimisation pushes to 60-65%, transforming operational efficiency.

Implementation Roadmap

Phase 1 | 0-30 Days

✓ Enable Drivers Licence for Kraken

✓ Enable Passport as fallback source

✓ Configure cascade: NatID → DL → PP → Manual

✓ Priority: FRA, NLD, GBR, DEU, CHE, CAN

Phase 2 | 30-60 Days

✓ Optimise US SSN stack (secondary sources)

✓ Activate Mobile Phone in EU

✓ Deploy regional verification profiles

✓ Address Belgium + Norway coverage

Phase 3 | 60-90 Days

✓ Expand to remaining 69 zero-pass countries

✓ Weekly pass rate monitoring dashboard

✓ Review & iterate based on results

✓ Explore eID integration (BEL, NLD, NOR)

Comments

No comments yet