Google's Helpful Content Questions, as a Checklist You Can Score

How to use this scorecard

1. Pick 10 pages: your biggest losers after an update, or your most important pages.
2. For each page, open the top 3 results that now outrank it in another tab.
3. Answer each question below with Yes (1 point) or No (0). The "warning sign" questions score the other way: a Yes there costs a point.
4. Google suggests asking someone you trust who is not linked to your site to answer too. Owners grade their own pages kindly.

Scoring on its own decides nothing. It tells you where a page is weakest compared with the pages beating it.

Content and quality (12 checks)

Expertise (4 checks)

People-first (5 checks)

Search-first warning signs (11 checks, Yes = minus 1)

Google lists these as signs to rethink how you make content:

• Made mainly to attract visits from search engines.
• Lots of content on many unrelated topics, hoping some of it ranks.
• Heavy automation used to cover many topics.
• Mostly summarizing what others say without adding much.
• Written because a topic is trending, not because your audience needs it.
• Leaves readers needing to search again for a better answer.
• Written to a word count because you heard Google prefers one. (Google says it has no preferred word count.)
• Entered a niche with no real expertise, mainly for search traffic.
• Promises an answer to a question that has none, such as an unconfirmed release date.
• Changed the date to look fresh without real changes.
• Added or removed lots of content mainly to make the site "seem fresh".

Who, How and Why (3 checks)

Google also asks you to look at:

• Who made the content. Is that clear to visitors, with a byline where people expect one, leading to more about the author? Google warns against made-up author profiles, such as AI headshots or false credentials.
• How it was made. For reviews and anything where method matters, show the work. If automation or AI did a large share, would a visitor expect to be told, and are you telling them?
• Why it exists. The main reason should be to help people who come to your site directly, not to rank. Google says using automation or AI "for the primary purpose of manipulating search rankings" breaks its spam policies.

How to read the scores

Add up the page's points (out of 24, after the warning sign deductions) and compare with the same score for the pages that outrank it.

Worked example (fictional)

brightsmile-dental.com scored its "teeth whitening cost" page after the May 2026 core update. It scored 11 of 24. It had no prices from the clinic itself (question 1), no named dentist (13 and 15), and it was written to 2,000 words because "long pages rank" (a warning sign). The top result, a clinic page with a real price table and a dentist's name, scored 19. The fix list wrote itself: add the clinic's own price ranges, name the reviewing dentist, cut the padding.

What this is not

Google's quality raters use a much longer document, the Search Quality Rater Guidelines. Raters give feedback on Google's systems but do not set rankings for any page. Google says that of experience, expertise, authoritativeness and trust (E-E-A-T), trust matters most.