Cake by VPBank builds a measurement foundation for Vietnam's fastest-growing digital bank

As Cake scaled acquisition across Meta, Google, TikTok, affiliate, and web-to-app channels, its in-house Digital Marketing & Growth team ran into a problem that was structural more than technical.

Before Adjust, campaign data was fragmented across advertising platforms, internal reports and app analytics; each platform applied its own attribution logic, and our team had to manually reconcile numbers before any budget decision. The digital-banking funnel is long, with an install often followed by registration, eKYC, account activation, deposit opening, loan approval or card issuance — sometimes days apart. Solving that required our team to design a common event framework first, then choose the measurement infrastructure that could support it.

Cake measures marketing performance across three layers — volume, quality and efficiency — specifically to stop the organization from scaling cheap traffic that never converts into real banking value.

• Volume KPIs: clicks, installs, registrations, eKYC starts/completions, account creation, applications, product-event volume.
• Quality KPIs: activated CASA accounts, first transactions, term-deposit openings and deposit amounts, loan approvals and disbursements, credit-card issuance rates & active rates, successful international transfers, retention and reattribution.
• Efficiency KPIs: click-to-install CVR, CPI, cost per activated user, CPA per application, CAC per disbursed loan or issued card, channel-level efficiency, optimization speed and reporting accuracy.

Through this framework, Cake has used Adjust's attribution, in-app event tracking, partner postbacks and raw-data exports to improve data measurement quality and reporting efficiency by roughly 20–50%, depending on product, campaign and channel — identifying performance gaps faster, cutting discrepancies between ad-platform data and actual banking outcomes, and sharpening budget-allocation calls made by our team.

Cake’s analysts used Adjust's event attribution and raw-data exports to uncover a significant Android-vs-iOS efficiency gap after account creation, then reallocated budget and re-optimized campaigns toward term-deposit openings rather than installs. For lending and credit-card campaigns, the same team-built approach lets us evaluate performance on approved, disbursed or issued outcomes — reducing the risk of overvaluing low-quality applications and giving us a much clearer, internally-validated view of true CAC.

Cake’s growth team built its Adjust setup around mobile attribution and in-app event measurement, Advanced Lifecycle Management, Multi-Touch Tracking, the Fraud Prevention Suite, web attribution, Audience Builder, Pulse, SpendWorks, and TrueLink. 

Rather than track installs, the team defined its own event set, covering eKYC completion, activated CASA accounts, term deposit openings, loan approvals and disbursements, and card issuance, and moved every channel comparison onto those events.

That change showed its value in the Term Deposit product line. Cake analysts used Adjust data to compare Android, iOS, and web performance at the CASA and term-deposit stages rather than at install, and found that cost per created account on iOS ran nearly 4x the Android level. The finding, produced and interpreted internally, gave the team grounds to shift budget toward Android, pull back inefficient iOS spend, and retest event mapping, targeting, and creative.

The rollout itself connects app analytics and Firebase events through GTM Mobile into Adjust, with event tokens and postbacks for each marketing platform configured and maintained by Cake's own engineers.

Cake's growth team now tracks performance across three layers it defined itself: volume (installs, registrations, eKYC starts and completions), quality (activated accounts, deposits, loan disbursements, card issuance, retention), and efficiency (CPI, cost per activated user, CAC per disbursed loan or issued card). Measured against that framework, Adjust's attribution, event tracking, partner postbacks, and raw data exports helped the team improve data measurement quality and reporting efficiency, closing gaps between ad-platform numbers and actual banking outcomes faster and sharpening its own budget calls.

Key outcomes since partnering with Adjust include:

• Improved marketing data measurement and reporting efficiency by roughly 20-50%, depending on product, campaign, and channel.
• Uncovered a nearly 4x cost gap between iOS and Android at the account-creation stage, redirecting budget toward more efficient channels.
• Supported customer growth of roughly 2x in 2025 and 3.5x in 2026, backed by a measurement layer the growth team owns and governs internally.