The project’s creator grew tired of manually copying Wordstat queries into Excel and built an automated pipeline instead. The first script simply sent requests to the free Bukvarix API, collected keyword data, and saved it to a file. That eliminated routine work, but Bukvarix data lagged by several months, forcing you to plan campaigns on stale frequencies.
To fix the outdated numbers, a second independent source was added: XMLRiver. XMLRiver returns the same metrics as Wordstat, but without captcha and almost in real time. The two channels run in parallel; Bukvarix quickly covers synonyms while XMLRiver guarantees current frequency counts. At the code level an Ensemble Voting mechanism compares the two values for each request and automatically selects the most consistent result.
The outcome is a system that delivers fresh, reconciled metrics, boosting campaign forecast accuracy by at least 15 percent. No extra data scientists were needed—the ready‑made Python script relies on standard libraries, API integration, and built‑in voting logic. An automatic retry mechanism shields you from temporary XMLRiver outages, and a simple heuristic for keyword difficulty filters thousands of queries without expensive third‑party services.
Why this matters: you get a tool that speeds up the creation of semantic sets and reduces the risk of budget waste caused by outdated data. The solution scales without heavy investment in analyst staff and can become a competitive edge when launching new advertising campaigns.