Search still begins with a modest hope: that a careful page can be found by the person who needed it. The first large web engine treated the link graph as a web of citations, and it found that the words inside a link often describe a page more faithfully than the page describes itself [1]. The ranking built on that idea, PageRank, gives a page weight when other weighty pages are willing to point to it [2].
A familiar hope is that more links are simply better. Structures built only to borrow that weight were already described as link spam in 2005, and the same study placed the repetition of query words through titles, body text, and anchors under the name of term spam [3]. Those moves can lift a page for a while, and while they do, a careful site can sit lower than it deserves [4]. There is a more encouraging result beside that warning. In a model of search advertising, optimization improved the unpaid results when a site’s usefulness to people moved with the value the site hoped to earn, and the ads then gave weaker sites less reason to dress themselves up [5]. In that model, visitors began with the organic result [5].
The words on the page still matter, though a density score is a poor way to say how. An experiment found that visibility rose when the words people search for showed up more often in the title, in the body, or in both [6]. A later reading of that work noted that using them in the title and the body beat optimizing the title alone [4]. We would keep the lesson and leave the quota behind. Those engines are not the ones people use now, and repeating a phrase past the point a reader would welcome is the stuffing that taxonomy called spam [3] [6]. What still feels sound is a title that says what the page is for, a passage that answers the question, and a link written by someone who has actually read you [1] [6].
There is a result in that advertising model that rarely survives the retelling. Optimization can leave the visitor better off and the stronger site better paid, while the search engine itself typically earns less, because the money drifts from the sponsored listing toward the unpaid one [5]. On their account, a rising tide of optimization is not a gift to every party at the table [5].
We hold the older papers as clues. Ranking systems move, and a finding from 1998 or 2005 is not a recipe for this year [1] [6]. When a tactic has to be hidden from the person reading, it is usually on its way to being hidden from the engine as well [3] [4].
What still feels sound?
- A title that says what the page is for.
- A passage that answers the question.
- A link written by someone who has actually read you.
How do those habits compare?
The papers did not score these habits the same way.
| Habit | What the research reported |
|---|---|
| Links built only to borrow weight | Described as link spam in 2005. |
| Repeating the query through the page | Described as term spam in the same taxonomy. |
| Search words in the title and the body | Visibility rose in a 2005 experiment. That is not a density quota for the engines people use now. |
| Optimization when usefulness moves with value | In a 2013 model, the unpaid results improved, and visitors began on the organic result. |
Are more links simply better?
Structures built only to borrow that weight were already described as link spam in 2005.
Does a keyword-density quota still decide visibility?
An older experiment saw visibility rise when search words appeared more often, and the quota itself is the part we would leave behind.
Can search optimization help the person searching?
In one model of search advertising, optimization improved the unpaid results when the site’s usefulness moved with the value it hoped to earn.
Should a paper from 1998 be used as this year’s recipe?
Ranking systems move, and a finding from 1998 or 2005 is a clue, not a recipe for this year.
References
- Brin, S., & Page, L. (1998). The anatomy of a large-scale hypertextual Web search engine. Computer Networks and ISDN Systems, 30(1–7), 107–117. https://doi.org/10.1016/S0169-7552(98)00110-X
- Page, L., Brin, S., Motwani, R., & Winograd, T. (1999). The PageRank citation ranking: Bringing order to the web. Technical Report, Stanford InfoLab. http://ilpubs.stanford.edu:8090/422/
- Gyöngyi, Z., & Garcia-Molina, H. (2005). Web spam taxonomy. Proceedings of the First International Workshop on Adversarial Information Retrieval on the Web. https://airweb.cse.lehigh.edu/2005/gyongyi.pdf
- Malaga, R. A. (2008). Worst practices in search engine optimization. Communications of the ACM, 51(12), 147–150. https://doi.org/10.1145/1409360.1409388
- Berman, R., & Katona, Z. (2013). The role of search engine optimization in search marketing. Marketing Science, 32(4), 644–651. https://doi.org/10.1287/mksc.2013.0783
- Zhang, J., & Dimitroff, A. (2005). The impact of webpage content characteristics on webpage visibility in search engine results (Part I). Information Processing & Management, 41(3), 665–690. https://doi.org/10.1016/j.ipm.2003.12.001