By Ayan Chakraborty, Founder, SBA Innovation Pvt Ltd, Kolkata. Last updated . Reviewed by Ayan Chakraborty.
How do I use offline conversion imports to optimise loan ads on Google?
Offline conversions let Google Ads learn from what happens after the form. When someone clicks your ad, capture the click identifier, GCLID, from the landing page URL and save it as a hidden field with the lead. When the lead reaches a later stage in your CRM or loan origination system, upload that stage to Google Ads with the click identifier, the conversion time and optionally a value.
Choose the stage with enough volume to train bidding. If approvals are too few, import an earlier outcome such as qualified lead or login, and move down the funnel as volume grows. Respect consent: collect and send identifiers only as your privacy notice states, and keep records. Then judge campaigns on cost per approved application, prune search terms by approval data, and expect the first benefits after enough imports have accumulated for the bid strategy to learn.
| Step | What you do | Watch-out |
|---|---|---|
| Capture | Store GCLID in a hidden form field | Lost if the URL parameter is stripped |
| Store | Save it against the lead in the CRM | Match one click ID to one lead |
| Choose event | Qualified lead, login or approval | Pick a stage with enough volume |
| Upload | Send click ID, conversion time and value | Use the original click time, not the upload time |
| Review | Compare cost per approval by campaign | Judge after enough imports have built up |
From our client work: pruning by approvals, not leads
In the MSME lender engagement, search terms, placements and audiences were pruned against approved-application data instead of lead volume, and 214 negative keywords were added. Over the engagement, cost per approved application fell from ₹4,880 to ₹3,020 with the monthly media budget held flat, alongside the funnel rebuild.
Method: Engagement 027: approvals counted at credit approval in the LOS; media budget held at ₹11.5L a month by agreement so the change is attributable to funnel work.
What did not work: Channel reporting that stopped at 'lead' looked healthy while media spend grew 40% in two quarters and approvals stayed flat.
Read case study 027 MSME lender funnel rebuild with footnotes.