How to Scrape a Web Page to a Google Sheet With Bardeen
Bardeen builds a scraper in about five clicks: open the page, hit the Scraper, click the elements you want, and it generates a repeatable automation that writes rows to Google Sheets. Free accounts get 100 credits a month, which covers a few hundred rows on a simple page before you need to pay.
Bardeen is a Chrome extension that lives in your browser tab rather than a server. That single design choice explains both its strength (it sees pages exactly as you do, including pages behind your login) and its weakness (nothing runs unless your browser is open).
What does the free plan allow?
100 credits a month on the free tier, with scheduled automations limited to daily rather than hourly runs. Paid plans start around $20 a month for 500 credits and hourly scheduling.
A scrape of one page costs a few credits. Enriching 100 rows with contact data can cost 100 credits on its own. That is the part people miss: the expensive operation is usually the enrichment step, not the scraping.
| Plan | Price | Credits/month | Scheduling |
|---|---|---|---|
| Free | $0 | 100 | Daily |
| Starter | ~$20/mo | 500 | Hourly |
| Teams | ~$60/mo | 2,000 | Hourly, shared |
How do you build the first scraper?
- Open the target page in Chrome and let it fully load, including lazy-loaded content.
- Click the Bardeen extension, choose New Automation, then Scraper.
- Click each field you want — title, price, rating — and name it. Bardeen highlights every matching element on the page so you can see immediately if the selector grabbed the wrong thing.
- Preview the table and fix any column showing null or repeated values.
- Add the write step: Google Sheets, pick a spreadsheet and a sheet tab.
- Run it once on the current page and check the rows landed correctly.
Step 3 is where the quality is decided. If a price renders as “$12.99/mo (billed yearly)” and you want the number, add a small transformation after, not during selection.
How do you make it paginate?
Add a pagination step with a next-page selector, and tell the scraper how many pages to walk. Most failures happen here.
Three things to check on the pagination selector:
- It must be a stable attribute, not a generated class like
css-1x9f2a - It must be present on every page including the last one (or the loop errors at the end)
- It must wait for the new content to load before extracting
A scraper that works on page one and returns duplicates for pages two onward has a load-wait problem, not a selector problem. Add a short delay step before extraction.
What about LinkedIn and login-gated pages?
This is the honest caveat. Bardeen can read pages you are logged into because it runs in your browser session, which means it can technically scrape LinkedIn profiles and similar gated sites.
That does not make it permitted. LinkedIn’s user agreement prohibits automated scraping, and accounts get restricted. Companies have been sued over scraping at scale. If your workflow touches a site with terms prohibiting it, the risk lands on you, not on Bardeen.
For public directories, company websites, and listings pages, this concern largely does not apply. Use the tool there.
Does the scraper break?
Yes, eventually. Any site redesign that changes the class names, the DOM order, or the pagination markup will break a scraper, and you find out when a scheduled run writes zero rows.
Three habits that reduce the pain:
- Add a row-count check — if a run writes fewer than 10 rows, something is off
- Take a screenshot on failure if the plan allows, or log the URL
- Keep the scraper small — one page type per automation, not five
Expect to fix a scraper every few months for an actively developed site. That maintenance cost is real and belongs in your decision.
How does it compare with a coding scraper?
A Python script with Playwright does the same job with better error handling and no credit system. It also needs someone who can write and maintain it, plus a server to run it on.
Bardeen wins on time to first result — minutes versus an afternoon — and on pages behind authentication, where a script needs session cookies you have to refresh.
For a one-off list of 200 companies, Bardeen is faster. For a daily job running for a year across 50 sites, a script is cheaper and more reliable.
Where does Bardeen fall short?
Your browser has to stay open for anything that is not a cloud-run schedule, and the free tier’s daily cap makes hourly monitoring a paid feature.
It also cannot handle heavy JavaScript-rendered pages reliably when the render takes more than a few seconds, and it has no headless mode of its own.
Who is this for?
Sales and research teams pulling public directory data into a sheet without engineering support. Founders doing market research at a scale where copy-paste would take a day.
Not for compliance-sensitive data teams, and not for anyone who needs a scraper running reliably at 3am without a browser session.
Quick Summary
- Bardeen builds a Google Sheets scraper in about five clicks, no code.
- Free tier: 100 credits a month with daily scheduling only.
- Paid plans start near $20/month for 500 credits and hourly runs.
- Scrapers break when sites redesign; add a row-count sanity check.
- Gated sites like LinkedIn are technically reachable, legally risky.
