Chapter 2 Quiz: The Ranking Algorithm

Twenty-four self-check questions. Mix of multiple choice and short answer. Answer key is collapsed at the bottom — try the whole set before you open it.

Multiple choice

  1. The claim "there are exactly 200 ranking factors, and here is the list" is best described as: a) An accurate summary of Google's public documentation b) Folklore built on an old, round, illustrative Google figure c) A leaked internal Google specification d) True in 2010 and still true today

  2. A ranking signal is best defined as: a) A lever you pull to complete an optimization checklist b) A measurable piece of information about a page, query, or context that ranking systems may use c) A guaranteed cause of higher rankings d) A third-party authority score

  3. Which of these did Google explicitly characterize as a lightweight ranking signal affecting well under 1% of queries? a) Relevance b) Links c) HTTPS d) Mobile-first indexing

  4. Google has described Core Web Vitals / page experience primarily as: a) The single most important ranking factor b) A tiebreaker between pages of similar relevance and quality c) Irrelevant to ranking d) A guaranteed path to position 1

  5. The single most important thing ranking rewards is: a) Word count b) Domain age c) Relevance to the searcher's intent d) Keyword density

  6. "Freshness is a ranking factor — yes or no?" is a malformed question because freshness is: a) Debunked entirely b) A query-dependent signal (decisive for some queries, irrelevant for others) c) The same as word count d) Confirmed to matter equally for every query

  7. Google stated it stopped using which tag for ranking, back in 2009? a) The title tag b) The viewport tag c) The meta keywords tag d) The canonical tag

  8. RankBrain (2015) is best described as: a) A tool webmasters install b) Google's first major machine-learning ranking system, which helps interpret query meaning c) A third-party rank tracker d) A penalty for thin content

  9. BERT's main contribution to Search (2019) was: a) Counting keywords more accurately b) Reading words in context to grasp meaning, moving ranking beyond string-matching c) Blocking JavaScript pages from the index d) Measuring page speed

  10. "Pages ranking #1 have more backlinks than page-two pages" is, by itself: a) Proof that building links causes ranking b) A correlation that could also reflect reverse causation and confounders c) A confirmed Google statement d) Evidence that links don't matter

  11. On the evidence ladder, "I've seen this fix work on my site" is: a) Rung 1 — confirmed by Google b) Rung 2 — strong correlation c) Rung 3 — practitioner experience (real but anecdotal) d) Rung 4 — pure speculation

  12. Which is a no-regret move under uncertainty? a) Buying bot clicks to raise CTR b) Adding a meta keywords tag c) Covering a topic thoroughly for the reader, whether or not it's a ranking factor d) Padding every page to 2,000 words

  13. Google's public position for years was that it does not use click-through rate directly as a ranking signal. What later complicated that position? a) A Google blog post confirming a "CTR dial" b) Testimony in the 2023 DOJ antitrust trial and a 2024 documentation leak indicating click-based systems (e.g., NavBoost) exist c) An academic paper from 1998 d) Nothing — the position never changed

  14. The reason even Google can't fully explain a single ranking is that: a) Google refuses to look b) Ranking is a system of machine-learned components whose behavior is emergent, not a readable rulebook c) The algorithm is a single simple formula they keep secret d) Rankings are assigned randomly

  15. "This study proves 1,890 words is the ideal length" commits which core error? a) It confuses correlation with causation (thoroughness, not length, is doing the work) b) It cites Google documentation c) It underestimates word count d) Nothing — it's correct

  16. MUM (2021) should be filed as: a) A confirmed, heavily weighted general ranking dial b) Announced and real, but with an unconfirmed and probably narrow role in live ranking c) A debunked myth d) A third-party SEO tool

Short answer

  1. Explain, in two sentences, the difference between a ranking signal and a ranking system, with one example of each.

  2. Google confirmed HTTPS as a ranking signal. Why is it still a mistake to spend heavily on "SEO" that is really just switching to HTTPS and expect rankings to jump?

  3. Give the three fast questions you run on any SEO claim before acting on it.

  4. Name the confounder that explains why "older domains rank better" is a correlation, not a causal factor.

  5. A page ranks #1 and yet its clicks fell about 30% this year. Give the most likely explanation from this chapter, and name the two chapters where the topic is developed.

  6. Why is "keyword stuffing" a real risk while "keyword density" is a myth? Distinguish them in one or two sentences.

  7. In one sentence, state the honest strength at which you should describe a well-run vendor correlation study: which evidence rung, and why not higher?

  8. A consultant says "bounce rate is a confirmed ranking factor." Sort the claim by tier and state what Google has actually said, in two sentences.

Answer key **Multiple choice:** 1-b, 2-b, 3-c, 4-b, 5-c, 6-b, 7-c, 8-b, 9-b, 10-b, 11-c, 12-c, 13-b, 14-b, 15-a, 16-b. **Short answer:** 17. A **signal** is a measurable input the ranking machinery may look at (e.g., the links pointing to a page); a **system** is a component that uses signals to do a job (e.g., RankBrain, which interprets query meaning). Signals feed systems; systems produce the ordering. 18. Because "confirmed" establishes only that the signal *exists*, not that it is *powerful*. Google itself called HTTPS lightweight; if the page is losing on relevance, HTTPS changes nothing. Do it for security and correctness, not as a rankings strategy. 19. (1) Which evidence rung is this claim on? (2) What observation would prove it wrong (is it falsifiable)? (3) Who benefits if I believe it (what is the source selling)? 20. Time: older domains have simply had more years to accumulate links, content, and authority — those are doing the work, not the domain's age itself. A ten-year-old empty domain ranks like nothing. 21. An **AI Overview** (Google's AI-generated answer atop the results) is likely answering the question on the SERP, so searchers get the answer without clicking — ranking well is necessary but no longer sufficient. Developed in **Chapter 13** (content response) and **Chapter 36** (its home). The "30%" is illustrative. 22. **Keyword stuffing** — cramming a term unnaturally — is a real spam signal that can hurt you. **Keyword density** — the belief in an ideal *percentage* of keyword use — is a myth; ranking systems read meaning, not a term-frequency ratio, so there is no target to hit. 23. **Rung 2 — strong correlation, not proven causal.** Not higher, because a study shows association, not causation; confounders and reverse causation remain open, and the publisher's incentive is rarely neutral. 24. **Rung 2 dressed as rung 1:** it's a correlation presented as a confirmed causal factor. Google has repeatedly said it does not use Google Analytics bounce rate as a ranking signal; relevant, useful pages both rank well and retain visitors, so relevance drives both (reverse of the claimed arrow).