When looking to understand how search engines work, it is useful firstly to consider what the purpose of a search engine is:
‘A search engine’s objective is to provide the user with links to the most appropriate pages on the internet which pertain to the query which they have typed in.’
Let’s imagine a user types in the following query: ‘how do search engines work’.
For now let’s take the internet out of the equation – imagine there’s no internet it’s easy if you try (sorry couldn’t resist). So, how might you go about researching something like this?
Firstly you might ask someone you know, respect and trust if they know of any good sources, for example:
- someone who teaches computer science
- someone who works for a search engine
- a librarian
Now having asked around hopefully you’ve been given a shortlist of sources – how would you move forward?Remember – you’re not just looking for any old answer – you are looking for the most appropriate answer…
You might begin to narrow down your search by sorting through your sources -
Start by looking for content with a title either the same or very similar to the query; then begin to read through some of the content to see whether or not they actually provide an answer to your query.
Probably by reading some extracts from your sources you can begin to narrow down your selections further and come up with a list of content which answers the query.
So here comes the hard part – which is the most appropriate resource?
Now obviously, you are not an expert; so it’s very tricky to decide which is the most appropriate – but what sort of things might you look for?
- How about how new / recent the source is?
- Newest doesn’t necessarily mean best, but theories change over time – recency could be a good measure.
- How about citations? Does the source refer to other sources?
- How well respected is the author? Are they considered an expert in their field?
Ultimately, Search Engines think like you…
Now, the process which you’ve been through really isn’t all that dissimilar to how a search engine works. A search engine is not an expert, and therefore no more or less ‘qualified’ to answer a query than you or I.
Instead a search engine works to try to provide the most appropriate answer by applying a complex algorithm which emulates a human’s decision process.
Search engine’s look at some of the following factors -
1. How well respected is the web page?
Search engines use links to decide how well respected a given web page is – they see links like votes – i.e. if people link to something it means they like the content. (NB obviously not all links are born equal – a link from a strong page on a university site carries more weight than your best friend linking to you from their blog).
2. Recency
When the page was created / last updated
3. How relevant does the content seem to be?
In order to understand what the page is about, the search engine will analyse the words on the page – e.g. how frequently they appear across the page, in headings, in titles etc
Now exactly how each search engine weights these factors and indeed precisely how they measure them is kept secret by the search engines, as understandably they need to protect their algorithms in order to protect the integrity of their results.
Nevertheless – hopefully this will haven given you some insight into how search engines work.
Agree, disagree; or if you’ve any questions comment below, or contact us.
Image credit Joel Bedford
