Our SEO Woes: May 2013

How Panda and Penguin hit a hand-built fashion directory — a first-hand account from May 2013, updated February 2014

Rather than lead off with a flowery introduction, this article jumps straight to the point: in May 2013, Google's algorithm updates hit Apparel Search hard, and left the site wondering what it had done wrong in regard to search engine optimization. It covers Google's reasoning as far as it could be understood, the site's own guesses about the cause, and the question of whether Google was punishing small business.

Since the beginning of its development, Apparel Search's efforts were strictly dedicated to building a website helpful to its viewers. Originally those efforts were rewarded: a few years before 2013, typing an apparel industry term into Google — "fashion news," "fashion jobs," "fashion designers," "clothing stores" — would put Apparel Search near the top of the first page, deservedly so in the site's view, having been first to market in many respects, and not just for a few terms. By May 2013 the site was rarely found even for the most evident search query for an apparel industry directory, and it was not pleased to see websites ranked above it that clearly did not deserve the ranking.

Early-2010s office desk with a desktop computer showing a downward traffic graph, printed website pages, a notepad, clothing catalogues and an Apparel Search desk plaque

Google's Side of the Story: Panda and Penguin

Truth be told, the site did not actually know Google's side of the story. As far as could be comprehended, Google's pet Panda and Penguin had decided to disrupt its traffic — and those two updates were running rampant across the entire internet, which made nailing down exactly why this website had been found so enjoyable to disrupt genuinely difficult.

Google Panda is a change to Google's search results ranking algorithm that was first released in February 2011. The change aimed to lower the rank of "low-quality sites" or "thin sites," and return higher-quality sites near the top of the search results. Google Penguin is a codename for a Google algorithm update that was first announced on April 24, 2012. The update was aimed at decreasing search engine rankings of websites that violate Google's Webmaster Guidelines by using now-declared black-hat SEO techniques involved in artificially increasing the ranking of a webpage by manipulating the number of links pointing to the page. (Moz maintains a history of when Google changed their algorithm for anyone tracing the timeline.)

Looking for an Explanation

To investigate the disruption to its traffic, the site visited Google's webmaster support and found the large blanket statements rather unhelpful. The Apparel Search site had over 55,000 pages at the time and had not used a cookie cutter to make them — each page created is unique — so generic information gave insufficient guidance for identifying and fixing site-specific issues. The next step, Google Webmaster Tools, helped identify a few minor issues but did not resolve the woes. The standard advice — make unique content, write for people not engines — was exactly what the site believed it had been doing from day one. Seeing that Google's educational resources had not cured the trouble, paid services such as Moz came under consideration; SEOmoz also provided many excellent free services.

The Site's Own Guesses

The pity party was not only about Apparel Search: many innocent and non-innocent websites were damaged when Google changed its algorithm, and Google's attempt to sweep across the mass of spam websites left more than a few innocent ones damaged in the wake — while also cutting back on genuinely spammy sites, which no one faulted it for attempting. The open question was whether this site deserved punishment, or whether some simple technical aspect was causing the reduced traffic. Working only from what was known about the site itself, four candidates stood out.

Dead links: yes, plenty, probably more than there should be — but with tens of thousands of pages it is not possible to keep a directory entirely clean at all times, because when companies go out of business or change their URL, they do not run it by the directory first. Unique content: the site is primarily unique, hand-developed content; the only pages without it are pages using an API that provides news and blog posts — and even those pages had countless hours of individual development behind them. Sitemaps: this looked like a big one — the site had not had a sitemap, set one up the week before this article, and as of May 2013 had not yet seen the benefit. And inbound links from its own sites: alongside Apparel Search, the company had developed other fashion websites such as the Fashion Newspaper and a few blogs, each for a unique purpose, cross-linked because linking a designer mention in a blog post to that designer's biography is natural and helpful to the viewer — but with tens of thousands of pages on each side, Google may have read helpful cross-linking as link farming.

Why the Punishment Felt Unfair

Three objections were raised. First, Google should not penalize an entire website if a certain section of the site has elements that do not meet its current standard — a huge site created properly should not be downgraded wholesale for wrong coding in one area. Second, the standard itself keeps changing: canonical tags, sitemaps, authorship and other creations arrive, publishers adjust, and an entirely different set of rules pops up the following day — a cycle that does not hurt SEO companies, who gain new clients, but does hurt small website publishers who lack the time for constant SEO work, the resources to hire consultants, or an inside web developer for every tiny adjustment. Third, the quality of apparel search results themselves seemed to have lowered: human judgment, knowing the apparel segment, said Apparel Search deserved to appear somewhere in the first pages of results for "apparel industry."

In companies named in this debate, the article weighed Microsoft and Bing against Google directly: for his own directory research, the author had for the previous few months turned to Bing as his primary search engine, finding its results better at that time — while noting that could change. The follow-up argument about one remedy Google was pushing appears in the article on the disavow process.

February 2014 Update: The Counter-Argument

An update added in February 2014 quoted a forum commenter's blunt counter-view. At one point, directories of topically related sites were useful because Google was not up to snuff with its results — ten years ago, even five. By 2014, in this view, definitely not: sites like Apparel Search compete directly with Google, and Google considers it a bad user experience when a user searches in Google and then has to search again in another site Google served up. Google wants its results to match the intent of the user's search — type in words, click the site that matches, done. If the user searches for "fashion jobs" or "clothing stores," Google wants to serve results that are fashion jobs or clothing stores, not a directory where the search continues; most users click Back immediately on landing on a directory, whereas a search for "apparel directory" would make the same user happy to find one. People had changed, and therefore Google had changed.

The commenter's conclusion: the most prominent value in a directory is giving visibility to unique and possibly high-quality sites that do not currently rank well in Google, but running the directory game unchanged was a losing battle against the world's largest search engine. The suggested path was a paradigm shift — reform the apparel site toward a review-based site where relevant businesses are showcased with interesting content a user would benefit from reading — because without something new and unique, the fight was unwinnable. The article lets that argument stand as the closing word of the 2013–2014 debate.

More SEO & Technology Guides

Common Questions

What were Google Panda and Google Penguin?

Google Panda was a change to Google's search results ranking algorithm first released in February 2011, aimed at lowering the rank of low-quality or thin sites and returning higher-quality sites near the top of the results. Google Penguin, first announced on April 24, 2012, aimed at decreasing the rankings of websites that violated Google's Webmaster Guidelines by using black-hat techniques that artificially increased a page's ranking by manipulating the number of links pointing to it.

Why did Apparel Search think it had been hit?

The site could only guess. Its candidates were dead links (hard to avoid across tens of thousands of directory pages when companies close or change URLs without notice), the lack of a sitemap (one was set up in May 2013, with no benefit yet seen), and inbound links from the company's own sister sites, which were cross-linked because the information was genuinely helpful to viewers but could have been read by Google as link farming.

What was the February 2014 counter-argument quoted on the page?

A forum commenter argued that directories once filled a real gap when Google's results were weaker, but that Google now wants to serve the thing searched for directly — a fashion job, a clothing store — rather than another directory where the search continues. On that view, a directory's remaining value is visibility for unique, high-quality sites that do not rank yet, and surviving meant a paradigm shift toward review and showcase content.

More from Apparel Search: Fashion Technology  |  Fashion Glossary  |  Fashion Companies  |  Fashion Terms
You are here: Apparel Search » Fashion Technology » Search Engine Optimization » Our SEO Woes, May 2013