Self-service onboarding for publishers
Defined the use cases and functional flows, prioritised the backlog with Engineering, and delivered a self-service path that shortened time-to-value for internal teams and external partners.
PO-01
Open to GameloftSummary
Featured attachment
Motivation Letter — Project Manager (BI), Gameloft
Product Ops — turning complex operations into products that deliver
Eleven years turning messy operations into shipped product. I own backlogs, write the use cases and functional flows behind them, and keep engineering, commercial and support pulling in the same direction — from PlayStation player support to a mobile monetisation platform handling 40Bn+ daily ad opportunities.
- 11+
- Years cross-functional
- 6
- Years gaming / ads
- EMEA+APAC
- Regions delivered
- 3
- Working languages
Highlighted epics
Fraud prevention programme
Set up a cross-functional claims process across finance, accounting and commercial, then partnered with data science on rules-based and ML detection — with incident response and stakeholder reporting.
Support platform & KPI layer
Owned Zendesk as a product surface: ticket taxonomy, SLAs, dashboards and forecasting used for budget and resource planning at leadership level.
PO-02
Delivery history
Mar 2025 — Jul 2026
Berlin
DoneOperations Manager
Upway GmbH
E-commerce marketplace · refurbished e-bikes (D2C, asset-heavy resale)
- Redesigned the European refurbishment-to-delivery workflow, lifting on-time order accuracy while unit volume grew — margin protected without adding headcount.
- Turned ad-hoc escalations into a prioritised, owned queue reviewed weekly with Sales, Marketing and Product, cutting recurring delays that were costing repeat purchases.
- Introduced a shared operational dashboard that replaced status meetings and gave leadership one view of delivery risk per market.
Sep 2019 — Feb 2025
Berlin
DoneTeam Lead, Offerwall Tech Ops
Digital Turbine / Fyber
Mobile ad-tech · rewarded advertising & app monetisation (revenue-share platform)
- Shipped self-service publisher onboarding — time from signature to first revenue down ~80%, with 95% of new accounts going live with zero support touches.
- Co-built rules-based and ML fraud detection with data science, avoiding a six-figure annual payout loss and protecting advertiser trust in the rewarded inventory.
- Ran delivery for a platform handling 40Bn+ daily ad opportunities: backlog owned in Jira, functional flows signed off before build, SLAs and KPIs reported to senior leadership.
- Rebuilt the support platform as a product surface — ticket taxonomy, SLAs and dashboards that cut incident resolution time by 40% and fed root causes into the roadmap.
Nov 2017 — Aug 2019
Berlin
DoneMerchant Operations Coordinator
SumUp GmbH
Fintech · SMB payments hardware & acquiring (transaction-fee model)
- Drove Salesforce adoption into daily workflows, removing manual tracking errors and giving Sales, Ops and CX one shared view of every merchant order — faster terminal activation, faster first transaction.
- Reduced order-related tickets through root-cause analysis, fixing logistics hand-offs instead of re-answering the same complaints.
Apr 2014 — Oct 2017
Berlin
DonePlayStation Client Operations
Sony Europe
Gaming · console platform, digital store & live services (EMEA retail + online)
- Escalated and reproduced in-game and account defects with specialised teams, shortening the loop between player-reported issues and shipped fixes.
- Coordinated EMEA retail logistics and system integrations for seasonal campaigns, protecting launch-day revenue across multiple markets.
PO-03
SDLC & BI projects
Custom application development · end-to-end SDLC
Custom application development across every phase of the SDLC — with BI delivery at its core.
Experience owning custom applications end to end: from documented requirements to post-launch tuning. The case below is BI work I led for fraud prevention at Digital Turbine — protecting advertiser spend and publisher payouts across mobile game monetisation.
01
Requirements
Gathered business and functional requirements from advertisers, publishers and internal finance — turned into documented use cases and acceptance criteria before any build.
02
Design
Co-designed the detection data model and the claims operating flow with data science and engineering, defining how rules and ML signals would be evaluated and audited.
03
Development
Owned the backlog that sequenced detection rules, the claims intake tooling and the reporting layer — prioritised by expected loss avoided, sprint by sprint.
04
Testing
Defined test scenarios for edge cases and false-positive limits with data science; validated rules against labelled traffic before promoting them to live payouts.
05
Deployment
Rolled out detection signals behind feature flags and shipped in fortnightly increments, so the payout pipeline was never at risk during release.
06
Maintenance
Owned the post-launch KPI layer: monitored fraud exposure monthly, tuned thresholds and fed root causes back into the roadmap as new backlog items.
BI project · Fraud Prevention — Digital Turbine
Advertisers & publishers on mobile games
On a rewarded advertising platform handling 40Bn+ daily ad opportunities, I delivered the BI layer that made fraud visible and actionable — for both the advertiser side (spend protection) and the publisher side (payout integrity).
Advertiser fraud detection suite
Rules-based + ML detection across rewarded video and offerwall inventory, protecting advertiser spend on mobile game campaigns.
- Rules engine
- ML signals
- SQL
- KPI dashboards
Outcome: Six-figure annual payout leakage avoided; fraudulent traffic blocked before payout, not post-hoc.
Publisher claims & payout integrity
Single-owner intake and audit trail for publisher disputes, integrating finance, accounting, commercial and engineering into one SLA-bound process.
- Cross-functional flow
- GDPR-by-design
- Audit trail
- Reporting
Outcome: Claim cycle time cut from ~3 weeks to ~5 days; monthly exposure reporting delivered in minutes vs. 2 days manual.
PO-04
Role fit
Role fit · Gameloft Project Manager — BI
What the role asks for, and where I have already done it.
- 01
Lead a transversal, high-impact project end to end
Ran multi-year delivery for a monetisation platform handling 40Bn+ daily ad opportunities, from scope definition to launch, across Engineering, Commercial, Finance and Support.
- 02
Coordinate internal resources and third-party vendors
Coordinated engineering squads, data science, external publishers and vendors across EMEA and APAC, keeping dependencies tracked as first-class plan items.
- 03
Deliver on time, in scope and on budget
Delivered self-service publisher onboarding that cut time-to-first-revenue by ~80% with no added headcount; scope changes traded openly, never silently.
- 04
Make sure requirements are properly documented
Wrote use cases, functional flows and acceptance criteria signed off by engineering before build — from actor to edge case.
- 05
BI, reporting and dashboards
Built the KPI and SLA reporting layer used by senior leadership for forecasting, budget and resource planning; SQL for Business Analytics certified (2023).
- 06
Measure performance, manage risk, escalate
Tracked delivery against SLA and KPI baselines, ran incident and escalation management, and reported risk per market weekly to leadership.
- 07
Agile project plans (Scrum / Kanban)
PMI-certified and Kanban-practised: planning, refinement, stand-ups, reviews and retros as the lead accountable for the outcome.
- 08
Gaming industry experience, Bucharest-ready
Six years on a mobile game monetisation platform plus three years inside Sony PlayStation player operations; based between Berlin and Bucharest.
PO-05
Backlog
Product Operations · backlog in action
Epics, user stories and measured outcomes — filtered your way.
Three product operations initiatives. Switch the lens to see how the work was framed, sliced and measured.
EPIC-01 · Publisher self-service onboarding
Product goal: remove humans from the critical path to first revenue
Publishers depended on manual hand-holding to go live on the Offerwall. I owned the operations backlog that turned onboarding into a repeatable product flow. — Digital Turbine / Fyber
OFW-114Done
As a publisher, I want to configure my Offerwall placement without a support ticket, so I can go live the same day.
Functional flow + acceptance criteria written with Engineering; sliced into 3 sprints.
Impact: Onboarding lead time cut from ~5 days to same-day (≈80% faster)
OFW-127Done
As a support agent, I want self-help documentation surfaced in-product, so repeat questions never reach the queue.
Knowledge base restructured around the top ticket drivers from Zendesk data.
Impact: ~35% fewer integration tickets per new publisher
OFW-151Done
As an account manager, I want visibility into integration status, so I can intervene before a partner stalls.
Prioritised above cosmetic requests after stakeholder review with Sales.
Impact: Stalled integrations spotted in <24h instead of at month-end review
EPIC-02 · Fraud detection & claims operations
Product goal: protect payouts with an auditable, single-owner process
Fraud claims crossed finance, accounting, commercial and engineering with no owner. I defined the operating process and the detection roadmap. — Digital Turbine / Fyber
FRD-08Done
As finance, I want a single intake for fraud claims, so every case has an owner and an audit trail.
Cross-functional flow designed and agreed with four departments before build.
Impact: Claim resolution time down from ~3 weeks to ~5 days
FRD-22Done
As the platform, I want rules-based and ML signals on suspicious traffic, so fraud is caught before payout.
Backlog co-prioritised with data science by expected loss avoided.
Impact: Six-figure annual payout leakage avoided; detection pre-payout, not post-hoc
FRD-40Done
As leadership, I want fraud KPIs in one dashboard, so exposure is visible each month.
Reporting layer shipped last, once the data model stabilised.
Impact: Monthly exposure reporting delivered in minutes vs. 2 days of manual collation
EPIC-03 · Merchant logistics visibility
Product goal: one source of truth for order state across ops, CX and 3PLs
Order tracking lived in spreadsheets and inboxes. The backlog focused on one thing: make status visible to everyone at once. — SumUp
MOP-05Done
As a merchant ops coordinator, I want Salesforce as the single source of truth, so manual tracking errors disappear.
Migration sequenced so daily operations never paused.
Impact: Manual tracking errors down ~90%; zero downtime during migration
MOP-17Done
As a CX agent, I want order state visible in the ticket, so I can answer without chasing logistics.
Prioritised using ticket-volume analysis by root cause.
Impact: Order-related tickets down ~30%; first-response time down ~40%
MOP-29Done
As a 3PL partner, I want a defined incident path, so exceptions get resolved inside SLA.
Agreed with external partners, then documented in Confluence.
Impact: Exception SLA compliance from ad-hoc to ~95% inside target
PO-06
KPI charts
Impact · measured
The KPIs behind the backlog, before and after delivery.
Every bar is a metric I owned, tracked in a dashboard and reported to leadership. Read the glossary below for exactly how each one is defined.
Publisher time-to-first-revenue
−80%Before~5 business days (manual)
AfterSame day (self-serve)
Fraud claim cycle time
−76%Before~3 weeks, no single owner
After~5 days, audited intake
Order-related support tickets
−30%BeforeBaseline volume
AfterAfter visibility shipped
Monthly exposure reporting
−98%Before~2 days manual collation
AfterMinutes, dashboard-driven
70%
Self-serve onboarding adoption
New publishers live with zero support touch
95%
Exceptions resolved inside SLA
3PL incident path vs. inbox escalation
88%
Backlog delivered per quarter
Committed scope shipped, 2-week cadence
PO-07
Results at a glance
Results at a glance
What changed for the business — in plain words.
Four outcomes, each with the problem, what I did and the number that moved. No jargon required.
Go-live in days, not weeks
New publishers used to wait on manual setup by the ops team. I shipped a self-service path, so they integrate and start earning on their own.
-80%time from signature to first revenue
95% of new accounts now go live without a single support touch.
Bad payouts stopped before they leave
Together with data science we turned manual claim checks into rules and ML signals that block suspicious payouts before money moves.
6-figureannual payout loss avoided
Claim decisions cut from weeks of e-mail chains to a tracked, days-long process.
Fewer tickets, faster answers
I treated the support platform as a product: clear ticket categories, SLAs and dashboards that showed which product gaps caused the noise.
-40%incident resolution time
Recurring ticket categories removed at the root instead of being re-answered.
The team ships what it commits
Backlog prioritised with engineering and commercial, scope agreed per two-week increment and reviewed openly in retros.
Bi-weeklypredictable delivery cadence
Stakeholders always knew what was landing next and what was traded away.
PO-08
Toolkit & education
Labels
- Backlog ownership & prioritisation
- Use cases & functional flows
- Agile & Kanban delivery
- Jira & Confluence
- Zendesk (owned as a product surface)
- Salesforce (operations adoption)
- SQL for analytics (certified 2023)
- KPI, SLA & dashboard reporting
- Mobile ad-tech / offerwall domain
- Stakeholder alignment (EMEA & APAC)
- Incident & escalation management
- Risk and timeline management
Education
- Project Management Certificate — PMI (2021)
- Brand & Product Management — IE Business School (2020)
- CX & Brand Management — University of the Arts London (2022)
- SQL for Business Analytics (2023) · Prompt Engineering (2025)
- B.A. Business Administration & Management — UAb
PO-09
Contact
Ready to run the BI project at Gameloft Bucharest.
Based in Bucharest and set up for remote delivery across time zones.
Languages
- Portuguese
- Native
- English
- Fluent (C1)
- Spanish
- Fluent (C1)
Attachment · reference letter
Reference letter — Fyber GmbH (Digital Turbine)
Issued Berlin, 28 Feb 2025 · signed by VP Engineering & HR Business Partner · covers Team Lead, Offerwall Tech Ops (2019–2025).
- Support operations leadership, SLAs, KPIs and Zendesk tooling ownership
- Fraud prevention strategy with data science, finance and commercial teams
- Reporting, forecasting and resource planning for senior leadership