Researching Web Sites

This article describes a technique that you could use to help you research

web sites. Researching web sites to see what your competition is up to

is cheaper than market research, doesn't take too long and can give you

a general overview of the current trends on the web.

I've used this - it works. Give it a try.

Step 1.

Find as many web sites whose goals and/or purpose are similar to those

of the web site you're working on.

If you use a search engine to find these sites, stop reading here and

go and bookmark as many as you can. Then come back and finish reading

this - it's important that you don't read the next paragraph in this article

before looking at these web sites.

No, I said go away and search for those sites.

Step 2.

Great, here's what I didn't want you to see before you searched for those

sites:

Note down all the keywords you used to search for these sites.

Hopefully, you've subconsciously saved yourself a lot of work. The top

five or six words you've used to search for your site should give you

a starting point when you insert your keywords into your meta-tags. I

didn't want you to think about your keywords while you were searching

as this could have biased your opinion of what to search for. As a side

note, you should do some usability testing and ask a sample of your target

audience to search for your web site (it doesn't matter if it's online

or not - you're only going to note down which words they typed into the

search engine). Back to the article…

Step 3.

Have a look through all the web sites you've found/selected to look at

and jot down a few notes about what you think of them. Here are some pointers

to help you:

• First impressions

Write down first impressions - did you find the whole thing easy to use?

Were you frustrated by the site? Did you stay longer than what you should

have done? Did you wander off track because something interesting caught

your eye? Did you think "what?" and leave straight away?

You only get one shot at a first impression so try to jot down as much

as possible.

• Design

Dog's bollocks or pig's ear? Like it? Hate it? Didn't notice it? Did

you say "Wow!"? Did you say "Urgh"? Simple, clean-looking,

cluttered, messy, horrendous? Professional looking? Amateurish? My 12

year old son could do better?

• Does it work?

Plagued by broken links? Does it work on your browser? On your Gran's

ageing copy of Netscape 2? Does the page render properly? Does that fancy

Flash animation load up? Tried it on a Mac yet? Go on, give it a shot;

you'll be surprised by the results.

• Content

Did you find what you're looking for? Is the information you expected

to be on the site actually on the site? Is there too much info? Too little?

Is it clear? Is the language (tone of voice) correct for the target audience?

Is it in the relevant place? Is it titled correctly?

• Navigation

Can you find the navigation elements? Do they work? Are they consistent?

Do they jump around the page? Is there too much choice? Too little choice?

Do you know where you are in the site at all times? Is there a "safety

net" link back to the homepage on every page? Are the link titles

descriptive enough? Are they obscure? Are they too long-winded?

• Functionality

Which features on the site really worked? Which features got in the way?

Any pointless gimmicks? Useful tools? Did you play around with a feature

for an hour? Did you enjoy it? Did you skip it because you didn't know

how it worked?

• Usability

Did you find the site easy to use? Did everything on the page make sense?

Did you feel you were always in control? Did you have to randomly click

everywhere in the hope that something would take you to another page?

Was the interface intuitive? (i.e. was it obvious from first glance or

did you have to think about how it worked?)

• Overall feel

Well, did you like it or not? Did it appeal to you? Anything you hated?

Loved? How many people have you sent the link to?

Step 4.

• Make notes
• Write as much or as little as you like
• You can do it one site at a time or include all of the sites you've
  looked at
• Make it informal
• Send it (electronically preferably - it's easier for everyone to add
  their comments) to other members of your team
• Send it to other teams if possible
• Let people go through the sites and add comments to yours
• Have a discussion (either through the document or face to face - do
  it at the pub if you prefer)
• Write down every point in the discussion no matter how insignificant
  it may seem
  
  If you have any ideas or suggestions on how you would have done things
  differently, write them down
• Share your thoughts and suggestions
• Let other people comment on them. (A good tip is to have people write
  their comments in differently coloured text, or different fonts if you
  are planning to print it out on a black and white printer.)

Step 5.

Now, leave it for a day or two. Get on with your work. Take a few days

off, visit your cousins down south.

Step 6.

Go back to the document, read through all the points. Visit the web sites

again to refresh your memory and to see examples of what people are referring

to.

Make bullet points of everything which people liked or disliked. Make

a note of any suggestions. Have a look at other web sites people may have

referred to.

You should now have something that will give you ideas on how to improve

your own web site.

Step 7.

Do the same exercise as above but this time on your own web site.

Step 8.

Now, take the lists of bullet points, compare them and write a set of

recommendations for the web site you're working on. What you're doing

here is looking for trends which users will identify with. You should

also see what works and what doesn't. Learning by your mistakes is great

but learning by your competitor's mistakes is cheaper.

What you do with these recommendations can depend on how your company

or department works. You may be writing these guidelines purely for your

own team to follow. If you need to justify a design decision to your client

(whether internal or external), you could refer to the web sites used

in your research. That way they can see how a design decision has been

implemented and whether it works or not.

---

Sources: Experience.

This article was first published on my brand spanking new IA site