An analytics system for an AdTech company
- Timeline
- MVP — 1 month
- in production 8+ years, since 2018
Task: bring 20+ advertising platforms into one system — reports by team and profit by campaign.
- Facebook Ads
- target.vk.ru
- TikTok Ads
- Google Ads
- Unity Ads
- AppLovin
- AppsFlyer
- Adjust.com
- Keitaro
- +12 more networks
ADTech platform
Universal connectors
One interface and one onboarding algorithm for any outside system
Permissions and limits
RBAC: people at different levels see and edit only their own data
Analytics and reporting
Flexible slices, exports and report setup
Realtime monitoring
Availability, rate limits and data volume
Media buyers
set up and run campaigns
Team leads
budgets and team results
Analysts
department-wide reporting
Management
P&L and buying efficiency
How it works
One interface and one algorithm for registering a third-party system
Every marketing system plugged into the platform — Facebook Ads, target.vk.ru, TikTok Ads — has its own way of connecting and its own way of authorising. What was needed was a layer of universal connectors: adding a network becomes straightforward and always the same, and the data coming back is uniform too, and plain to work with.
RBAC: employees at different levels see and edit only their own data
What the system shows is split by level of access, for four groups:
- Media buyers — the rank and file of the department. They set up campaigns from their own account and read their own figures, and nothing beyond that.
- Team leads — read the figures for their own team of buyers.
- Analysts — reach the wider financial data, with no right to edit it.
- Management — full access to every financial figure, but no ability to edit campaigns.
Flexible slices, exports and report configuration
The system is built around a flexible set of operations:
- data selection
- filtering
- sorting
- grouping
That scheme is built into the architecture itself, so each new dashboard, section after section, takes little time to put together, and the interface stays self-evident to the person using it.
Availability, rate limits and data volume
The system leans hard on heavy data processing, so it needs monitoring that can be seen through. Every network plugged into it brings its own connection limits, its own rate limits and its own set of connection errors. Each one is wired into the monitoring so those problems surface in good time, without anyone having to go looking for them.
WHO USES IT — media buyers run campaigns · team leads watch budgets and team results · analysts produce the department’s consolidated reporting · management sees P&L and buying efficiency. Facebook Ads, target.vk.ru, TikTok Ads, Google Ads, Unity Ads, AppLovin, AppsFlyer, Adjust, Keitaro and 12 more platforms.
Results
- 30 → 500
- growth of the department’s headcount
- +300%
- department profit
- 20+
- advertising platforms
Fast onboarding
Media buyers work to the same metrics from their first days
Campaigns under control
The efficiency of each buyer’s setup is visible, and results compare across any slice
Stack
PHP · Symfony · MySQL · ClickHouse · React · Docker