Chapter 2 Exercises: The Ranking Algorithm

Work these with the chapter open. Items marked † have worked solutions in the answers appendix; the rest are for discussion or self-check. There are no answers in this file on purpose — struggle a little first. Free tools you'll use: View Page Source (any browser), a site: search, and Google Search Console (if you have it). None of these exercises asks you to write code.

A. Recall — get the vocabulary exact

  1. † In your own words, define a ranking signal, a ranking factor, and a ranking system, and give one concrete example of each. Then explain, in one sentence, how the three relate.
  2. Define a query-dependent signal. Give two queries for which freshness is decisive and two for which it is nearly irrelevant.
  3. † List five ranking signals Google has actually confirmed it uses. Beside each, write one word — "big," "medium," or "light" — for how much it seems to weigh, and be ready to defend it.
  4. When Google announced HTTPS as a ranking signal, how did it characterize the signal's strength and reach? Why does that characterization matter?
  5. † Name RankBrain, BERT, and MUM, give the year each is associated with, and describe in one line what each contributed to how ranking works.
  6. State the four rungs of the evidence ladder, from strongest to weakest, with a one-line example of each.

B. Applied reasoning — use the ideas

  1. † Explain why "it's a confirmed ranking factor" is not the same claim as "improving it will lift my rankings." Use HTTPS or Core Web Vitals to make the point concrete.
  2. † A study finds that pages ranking #1 have far more backlinks than pages ranking #10. Explain reverse causation and how it could account for much of that finding.
  3. Why can't even Google's own engineers hand you a sentence of the form "page A beats page B for this query because of factors X, Y, and Z, weighted thus"? Tie your answer to machine learning.
  4. † A colleague says: "Google rewards quality, and if your quality page didn't rank, it just wasn't really quality." What is wrong with this claim as a piece of knowledge? Name the flaw.
  5. Explain the idea of a no-regret move. Why do "earn genuine engagement" and "cover the topic thoroughly" qualify, while "manipulate click-through rate with bots" does not?
  6. Two SEOs argue: one says "freshness is a ranking factor," the other says "no it isn't." Explain why the question, as posed, is malformed — and give the honest answer.
  7. Distinguish keyword density (a myth) from keyword stuffing (a real risk). Why does modern language understanding make the first idea obsolete?
  8. † "Long pages rank better" and "old domains rank better" are both real correlations that mislead. For each, name the confounder — the third thing that actually drives the ranking — and the wasted action the myth encourages.

C. Diagnose why this page isn't ranking

  1. † A page is indexed, loads fast, is served over HTTPS, is mobile-friendly, runs 2,500 words, and hits a tidy 2% "keyword density" for its target term — and it ranks #14. State the single most likely real problem, and which pipeline stage (from Chapter 1) it lives at.
  2. Rivertown's "AC Repair — Northgate" page is near-identical to its "AC Repair — Cedar Hills," "— Westbrook," "— Millhaven," and "— Rivertown" pages (the city name swapped in). Which confirmed signal is it failing, and explain precisely why "add more words to each page" will not fix it.

D. Audit your own site (free tools)

  1. † Open your own homepage, right-click, choose View Page Source, and search the raw code for keywords, viewport, and https. Report what you find for each and interpret it: which is a debunked "factor," which are confirmed signals, and what (if anything) you'd change.
  2. In Google Search Console (or, lacking it, a site:yourdomain search), pick one page and find its average position. Explain why that single number is not "your rank," drawing on Chapter 1's point that there is no universal ranking.

E. Research and sort real claims

  1. † Find one "SEO ranking factors" article online (there are thousands). Take its first ten claims and sort each into confirmed / likely / debunked / speculation. Note how many you can't confidently place — and why that difficulty is the whole point of this chapter.
  2. Here are three claims about title tags. Sort each by evidence tier before deciding what to do: (a) "Google uses the title to help understand what a page is about." (b) "You must put the exact keyword first in the title or you won't rank." (c) "Titles between 50 and 60 characters rank best."
  3. † Here is a real title tag: Services | Rivertown Home Services. First, sort the claim "putting the keyword in the title is a confirmed ranking factor" by tier. Then rewrite the title for someone searching furnace repair Northgate, and state honestly what evidence supports your rewrite and what it does not promise.
  4. Find a vendor correlation study about ranking. Identify (a) the correlation it reports, (b) a plausible confounder, and (c) whether reverse causation is possible. Then state what, if anything, you'd do differently because of it.

F. Read the report

  1. † A vendor chart shows: "Pages in position 1 have, on average, three times as many referring domains as pages in position 10. Conclusion: backlinks are the #1 ranking factor." Write the four honest reads: what it shows, what it doesn't, the move it justifies, and the lesson.
  2. A Search Console-style note reads: "Query 'furnace not igniting' — 9,000 impressions, 90 clicks (1% CTR), average position 8.2." A teammate says "our click-through rate is a ranking factor, so let's boost CTR to climb." Interpret the data and evaluate the teammate's reasoning.

G. Spot the myth

  1. † "Google rewards sites that publish something every single day." Myth or fact? State the evidence and what (if anything) is true nearby.
  2. "Our domain is twelve years old, so we're set for rankings." Myth or fact? What is actually doing the work in the correlation that fuels this belief?
  3. † "This study proves that 1,890 words is the ideal blog-post length." Spot every distinct error in that sentence — there are at least four.

H. Ethics and risk

  1. † A vendor offers a "CTR manipulation" service: networks of real-looking users clicking your result to raise its rankings. Evaluate it on two axes — does it work (evidence), and is it wise (risk/ethics).
  2. Your manager asks you to add a <meta name="keywords"> tag stuffed with competitors' brand names. Sort the request by both evidence (will it help?) and ethics/risk (should you?), and draft the two sentences you'd say back.

I. Rivertown Strategy-File extension

  1. † Take any three items from Rivertown's "proposed fixes, sorted by evidence" figure (§Strategy File) and, for each, write the single honest sentence you'd tell owner Tony about how much it will move rankings — without ever promising a ranking.
  2. Tony read online that "E-E-A-T is a ranking factor." Using only this chapter, explain what is true and what is misleading about that sentence, and name the chapter where the full answer lives.
  3. † Invent one more "fix" a home-services owner might plausibly hear from a friend, a forum, or a cold-calling agency. Sort it into confirmed / likely / folklore, and justify the tier in two sentences.