Previously, I asked the question of when it was appropriate to add within-page links to the top of a page and if there were any specifc guidelines concerning this.
The feedback from that post gave me food for thought and so I thought I’d take it one step further and actually come up with a calculation for when to add these lists of links to the top of your page.
This calculation takes into account the various factors that come into play in determining whether a web page is ‘long’ or not, such as screen resolution, word count, number of pages to scroll and so on.
Data
Here’s the data you’ll need to make the calculation:
- The word count for your article.
- The screen resolution (height, not width) of your target audience.
- The number of screens your page takes up at this resolution.
- The percentage of your total audience that your target audience constitutes.
Calculation
The calculation is as follows:
(Word count × No. of screens) ÷ Screen resolution × Percentage of total audience = Score
That’s it! Just as Pythagoras would have wanted.
Example
Let’s take an imaginary article for this site as an example. Here’s my data:
- Word count = 750
- Screen resolution (height) = 768
- Number of screens = 3
- Percentage of total audience = 31%
Thus, the calculation will look like this:
(750 × 3) ÷ 768 × 31% = 0.9
How to Interpret
The score will likely be somewhere between 0.5 and 3. Here’s how to interpret that score:
- 0.0 – 1.0 : no need to add jump links
- 1.01 – 1.5 : probably don’t need jump links
- 1.51 – 2.0 : probably do need jump links
- 2.01 and up : add jump links
Thus, in the above example, a score of 0.9 means there would be no need to add jump links to the top of the page.
Try it For Yourself
Thanks to the ‘magic’ of -Google Spreadsheets- Zoho Sheet, I’ve created an an editable example for you to play around with.
Parting Thoughts
I was going to add an additional variable based on the ratio of the number of characters to words in the title of the page, but then decided the formula was silly enough as it is.
Great post,
Maybe you can improve the results interpretation.
It’s frustating to have a range from 1.01 to 2.00 as result.
What does means exactly the “probably” word there.
How about to have:
0.0 – 2.0 : no need to add jump links
2.01 and up : add jump links
or
0.0 – 1.0 : no need to add jump links
1.01 – 1.5 : IF(criteria) don’t need jump links
1.51 – 2.0 : IF(criteria) do need jump links
2.01 and up : add jump links
PS: I’m confuse! My access is denied to your editable example at Google Spreadsheets 😉
*Maujor* – sorry that you couldn’t access the editable example. It looks like you have to be logged into your Google account in order to do so.
Despite the ‘seeming precision’ of my calculation, the reason I added ‘probably’ is that within a certain range (1.0 – 2.0) it can’t tell you for sure whether you need to add jump links or not.
Other factors need to be taken into account such as the type of article, the audience demographics and so on. All it can tell you is that at the lower end of the range you would likely not need to use them and at the higher end you probably should.
Perhaps I’ll create a version two of the calculation that takes these other factors into account. There are a couple of others to be considered too that I haven’t mentioned – can you guess what they are?
I’m trying to access your example:
We’re sorry.
It looks like you don’t have access to this spreadsheet.
But I am logged in into my Google Spreadsheets account 😉
Christian – Thanks for explanation. Sorry, I’m not an expert in this subject, so can’t help you pointing others factors to take in account or even some suggestions.
Yes, I was logged into my Google account, but access is denied.
___________________________________________________
The article is great, I’ve never saw something like it before. The Watson Formula is very useful! I’d like to share it with my readers, so I sent you an email asking for your permission to translate the article to Brazilian-Portuguese.
*nodh* – sorry about the Google Spreadsheets problem. I’ll try to recreate the example with another online spreadsheet program (should be up later today).
*Maujor* – be my guest. Feel free to develop it further as well. My own personal feeling is that the _Watson Formula_ is not nearly complicated enough. 😉
By the way, the other factors to take into account are line length (more words per line = shorter pages) and font size (larger fonts = longer pages). I’ve assumed a typical line length of about 10 words and a ‘normal’ font size.
Christian –
The Spreadsheet hosted at Zoho is working like a charm. Thanks.
For now i’ll translate the Watson Formula article and ask for feedback from my readers. I intend to make some experiments with Brazilian-Portuguese articles.
This is AWESOME. I tend to hate jump links, so I try to keep them to a minimum – scrolling, too. What are your thoughts on page length in general…?
Amy Eliot
*Amy* – I don’t have a definitive answer to that. It may sound like a cop out, but I think that a page should be as long as it needs to be.
As long as the content is well written and presented appropriately for the web then I’m not so concerned about page length.
My own usability testing has not shown that users are anti-long pages either.
However, if you have created a long page, I do think it’s worth taking some extra time to ensure that is going to be usable.
For example, if your page is long, you probably don’t want to have your most important content at the bottom.
There may be times where it is more appropriate to break up a single long page into a ‘section’ of several shorter pages.
Your audience makeup plays into this as well. For instance, I recall that older people have more trouble scrolling and tend to prefer shorter pages.