Case Study 2 — JCPenney, 2011: When Bought Links Ranked #1 for Everything, Then Didn't

A complementary angle. Case Study 1 showed the link signal working as designed — the crowd's independent judgment, harnessed. This one shows the signal being forged at industrial scale by a company large enough to know better, and Google reversing it in a single stroke. It is the clearest public illustration of the chapter's hardest lesson: buying links is detectable, the punishment is retroactive, and scale buys scrutiny, not immunity.

Type: Real, public, Tier-1. The events were documented in The New York Times in February 2011 ("The Dirty Little Secrets of Search," by David Segal) and widely confirmed afterward. Specific ranking positions and the outcome are drawn from that public reporting; where exact figures are unknown, this study says so and invents nothing.

Background: a retailer that ranked #1 for almost everything

In late 2010, if you searched Google for a startling range of lucrative shopping terms — "dresses," "bedding," "area rugs," "skinny jeans," "grommet top curtains," "furniture," and many more — one result appeared at or near the very top again and again: JCPenney, the large American department-store chain. For a company that sells nearly everything, ranking #1 for nearly everything is a windfall, especially heading into the holiday shopping season, when those top positions are worth an enormous amount of traffic and revenue.

It was too consistent to be organic, and a New York Times reporter, working with an SEO expert who analyzed the link profile, set out to find why. What they found is the textbook case §22.6 describes.

The investigation found that JCPenney's pages were being pushed up the rankings by thousands of backlinks from unrelated websites, many of them low-quality or effectively dormant sites, carrying keyword-rich anchor text pointing at JCPenney category pages. A site about, for example, nuclear engineering or classic cars or casinos would carry a link reading "dresses" or "bedding" aimed at JCPenney. There was no editorial reason for those links to exist — no relevance, no independent judgment, nothing the 1998 paper (Case Study 1) would recognize as a real citation. They existed to move rankings.

Read against Chapter 22, the profile was a confession written in every dimension at once:

FIGURE C2.1 — "Why the profile screamed manipulation"          [after NYT reporting, 2011]
  DIMENSION (§22.3/§22.6)   WHAT A NATURAL PROFILE LOOKS LIKE     WHAT JCPENNEY'S SHOWED
  Relevance                 Links from topically related sites    Links from thousands of unrelated sites
  Anchor text               Varied (brand, URL, descriptive)      Repeated exact commercial money-phrases
  Editorial intent          Freely given because content earned   No editorial reason to link at all
  Placement / source        In-content, real publications         Low-quality, often link-only pages
  Pattern                   Organic, uneven growth                A vast, uniform, purpose-built footprint

Notice that no single link was the problem; the pattern was. This is the key detection idea from §22.6: manipulation is visible in aggregate — velocity, footprint, irrelevance, and anchor over-optimization — even when any one link looks unremarkable. A human investigator saw it, and Google's systems are built to see the same shapes at scale.

🔎 How Search Sees It The links "worked" for a while, and that is the trap, not the exception. A forged link profile can lift rankings right up until the moment it is reassessed — by a spam-system refresh, a Penguin-style update, or, as here, a manual review prompted by evidence. The gap between "it's working" and "it collapsed" is exactly the retroactive-detection point from §22.6: JCPenney was not getting away with it during the good months; it was accumulating a liability with a delayed fuse, and the fuse was lit the moment Google looked.

What it shows

When the Times presented its findings, Google confirmed it viewed the activity as a violation of its guidelines and took manual action against JCPenney. (Manual actions — a human penalty applied by Google's web-spam team, distinct from an automatic algorithmic demotion — are the subject of Chapter 26.) JCPenney's rankings for the affected terms fell sharply and immediately; positions that had been #1 dropped far down, some by dozens of places, erasing the visibility the scheme had manufactured.

Three lessons land hard:

  1. Scale is not safety. A national brand with a large budget was more exposed, not less — high rankings across thousands of commercial terms are conspicuous, and conspicuous invites scrutiny. The idea that a big, "respectable" company is too important to be penalized is exactly backwards.
  2. The punishment targets the manipulation, not the company's legitimacy. JCPenney is a real, substantial retailer with plenty of genuine authority; Google did not dispute that. It removed the forged advantage. That is the system working as designed: devalue the fake votes, keep the real ones.
  3. The rise and the fall are the same event, seen twice. Everything that made the rankings possible — thousands of irrelevant, keyword-anchored, purpose-built links — is exactly what made the penalty inevitable once anyone looked. The tactic contains its own reversal.

Outcome

JCPenney's affected rankings dropped precipitously after Google's action, at a commercially painful time. The company distanced itself from the tactics and parted ways with the search firm involved, and — as is typical — recovery from a manual action was a slower, more painful process than the fall (the reconsideration path is a Chapter 26 topic). The episode became one of the most cited public examples of a paid-link penalty, precisely because the mechanism was so clearly documented: you could see the forged profile, see it work, and see Google undo it.

It is worth stating what this case is not. It is not evidence that "links are dangerous" — the opposite of its lesson. Links from real, relevant sites, freely given, are the strongest asset in off-page SEO. What is dangerous is forging them. JCPenney had the brand, the inventory, and the genuine authority to earn real links; it rented fake ones instead, and rented authority gets repossessed.

The lesson

Manufactured authority is a loan against your rankings, and Google is the lender that can call it at any time. The JCPenney case is §22.6 made real: the scheme is detectable because manipulation leaves an aggregate footprint; the enforcement is retroactive, arriving after the "success"; and the downside — sudden, severe, and slow to repair — dwarfs the temporary upside. The durable path was available to JCPenney the whole time and is available to Rivertown now: earn real citations by being worth citing (Chapters 23–24), and never build the business on links that can be taken back.

Discussion questions

  1. The case argues that JCPenney's high rankings and its penalty were "the same event, seen twice." Explain what that means using the §22.3/§22.6 fingerprints, and why the tactic essentially guarantees its own reversal.
  2. Google removed the forged advantage but did not dispute JCPenney's genuine authority. Why is that distinction important, and how does it reflect the link signal "working as designed"?
  3. "A brand this big is too important for Google to penalize." Use this case to refute that belief, and explain why scale can increase exposure.
  4. The links "worked" for months before the collapse. Explain, to a client tempted by a link vendor's short-term results, why early success is evidence of a delayed liability, not of safety.
  5. Contrast the remedy here (a manual action, requiring cleanup and reconsideration — Chapter 26) with simple algorithmic devaluation (the links quietly stop counting). Why does the type of enforcement change how a business should respond — and why should neither outcome tempt you to try the tactic?