Case study 4 · Lead filtering workflow for a Singapore lead generation company
The problem
Our client in Singapore generates leads for their own clients. Their challenge was the volume: thousands of leads, each with many attributes, stored in Google Sheets with columns that kept growing longer. Filtering them by hand was extremely time consuming.
The obvious fix was to hand all the data to AI and let it do the work. In practice, that approach broke down:
- Inaccurate output. With so much data, the AI struggled to retrieve the specific records it was asked for, and the results were often flawed.
- Runaway costs. With more than 3,000 leads, sending each one through an AI API made the cost of processing extremely high.
What we built
Instead of forcing AI to do everything, we engineered a workflow that uses the right tool for each job.
- Google Apps Script and macros read through each lead efficiently and directly inside the spreadsheet, so filtering is precise and repeatable.
- AI only where it adds value. By handling the heavy, structured work with scripts, the workflow avoids paying to send every single lead through an API.
- A path to scale. We also advised the client to move data of this size to a platform built for it, such as SQL or Supabase, so their lead database can keep growing without slowing them down.
The result
The client can now filter thousands of leads efficiently, with dependable output instead of AI guesswork, and at a much lower cost than processing every lead through an AI API.
The lesson is simple: a right approach will always reduce cost. AI is powerful, but it delivers the most value when it is part of a well-designed workflow rather than the whole solution.
Is this you?
If your team is drowning in spreadsheets, or your AI bills are growing faster than your results, let us review your workflow. We will help you find the approach that is accurate, affordable and ready to scale.