Data Sync Engine

Lyra’s Data Sync Engine automatically synchronizes campaign, keyword, search term, and performance data from all connected Google Ads accounts, maintaining a current and complete local data store that powers every Lyra tool and analysis feature.

Key Takeaways

• Automated daily sync keeps all data current without manual intervention
• UPSERT logic prevents duplicate records and handles data updates cleanly
• Backfill recovery automatically fills gaps from missed sync windows
• Foundation for all Lyra tools — every analysis and recommendation relies on synced data

The Problem

Google Ads data lives in Google’s systems. Accessing it for analysis, reporting, or optimization requires API calls that are slow, rate-limited, and constrained by query complexity. The consequences:

• Slow tool performance — Tools that query Google Ads in real-time are limited by API response latency, which can be seconds per request
• API quota pressure — Every tool, report, and analysis that directly queries the API consumes quota, creating contention across features
• Data availability gaps — Without local storage, data is only available when the API is responsive. Outages or rate limiting block analysis entirely
• Historical depth limits — Google Ads API provides limited lookback windows for some data types, particularly search terms

Without a synchronization layer, each Lyra feature would independently query Google Ads, resulting in redundant API calls, inconsistent data across features (different tools seeing different snapshots), and fragile performance dependent on API availability.

How Lyra Solves It

The Data Sync Engine runs as a scheduled background process that maintains a local mirror of your Google Ads data:

The sync architecture uses several techniques for reliability:

• UPSERT logic — Each record is inserted or updated based on its unique key. This prevents duplicates when a sync runs multiple times and handles cases where Google updates historical data retroactively.
• Incremental sync — After initial full sync, subsequent runs focus on changed data and new date ranges, minimizing API usage.
• Backfill recovery — If a sync fails mid-run, the engine tracks which date ranges were successfully processed. The next run automatically identifies and fills gaps.
• Parallel processing — Multiple accounts are synced concurrently, reducing total sync time for large portfolios.

Data freshness monitoring provides visibility into sync health:

• Last sync timestamp per account and data type
• Staleness alerts when data exceeds the expected refresh window
• Sync history showing success, failure, and backfill events
• API quota usage tracking to prevent quota exhaustion

Use Cases

Always-current analytics. Every Lyra dashboard, report, and tool operates on synced data. The Data Sync Engine ensures that when you open any analysis view, the data reflects the most recent sync cycle without waiting for live API queries.

Historical data preservation. Google Ads API limits lookback windows for certain data types. The sync engine captures and stores this data daily, building a historical archive that extends beyond Google’s default retention period.

Multi-account portfolio management. For agencies managing 60+ accounts, the parallel sync engine ensures all accounts are updated within the daily sync window. Account managers start each day with fresh data across their entire portfolio.