What attribution settings should I use for offline conversions imported into Google Ads or LinkedIn Ads?
Quick Answer: Use an attribution window that matches your real sales cycle, then keep it consistent across your CRM and ad platforms. For most B2B offline imports, a 30–90 day click window is the practical default; use shorter windows for high-volume, fast-close offers and longer windows only when you have enough data to justify the extra lookback. Avoid mixing models midstream, because changing windows, interaction types, or source mappings will distort optimization signals and make Google Ads or LinkedIn Ads learn from inconsistent conversion data.
Detailed Explanation
For offline conversions imported into Google Ads or LinkedIn Ads, the right attribution settings are the ones that align the platform’s conversion credit with your actual customer journey and CRM event timing. In practice, that means defining a single source of truth for the conversion event, using a stable lookback window that reflects your average time-to-close, and importing only the milestone that truly matters to bidding—typically qualified opportunity, SQL, or closed-won, not every low-intent lead touch. If your pipeline is long and multi-touch, a longer window can help capture more eligible ad interactions, but it should still be conservative enough to prevent inflated credit from stale clicks or impressions. The key is consistency: one event definition, one attribution rule set, and one mapping between CRM timestamps and ad platform identifiers so the algorithm can optimize on clean, comparable signals.
Key Technical Drivers
- Set the offline conversion window to match median sales cycle length: 30 days for short-cycle, 60–90 days for typical B2B, and only extend beyond 90 days if you have strong volume and stable close-rate data.
- Import the highest-value milestone you want the platform to optimize toward—usually SQL, opportunity created, or closed-won—and deduplicate by transaction ID so each deal is counted once.
- Keep attribution model, lookback window, and source mapping identical across Google Ads and LinkedIn Ads where possible; then evaluate performance using one consistent reporting layer in your CRM or warehouse.