Fine-tuning searches.. The better way!
So. You are making a search and you don’t get the results you want, next to that, if you want to fine-tune them so you do get the results you want and you can’t do it in such a way that is user-friendly and easy. How come?
Well, there can be a number issues, both on the user (consumer) side, and on the client (web shop) side.
User:
- Search is to general, for example, searching for an electronic appliance in an electronic appliances web shop should get you pretty much all the stuff in the shop.
- Search is to specific and user gets incorrect results
- Search isn’t correctly spelled and user gets incorrect results
Client:
- The products aren’t categorized correctly (the web shops tries to show the user more results so there are more categories then necessary)
- You can only specify how you want to sort things using drop down menus
- When you sort things using the already annoying drop downs, you still get all the rest of the results (the act of specifying the most important feature of your purchase doesn’t eliminate part of the product range)
- You have to specify what range of price you use by sliders or by typing (which imho isn’t that user-friendly)
- Comparing results is very difficult and takes a lot of “forward / previous” browser action
So, how do we go about fixing these things? Why don’t we discuss them peace by peace.
P: Search is to general, for example, searching for an electronic appliance in an electronic appliances shop should get you pretty much all the stuff in the shop.
A: Suggestion based searching where the user gets a warning when they exceed a certain amount of hits (a 100 or something). When they do get that warning, the user is presented with a tag-cloud in which they can choose how they want to make their search more specific.
P: Search is to specific and user gets incorrect results
A: This can be dealt with the same way as with the to general searches. The user gets a warning when his hits are blow (lets say) 3. Ultimately (depending on how many products you have in the shop) you want to have something like 25 to 50 products to choose from which contain all price ranges etc.
P: Search isn’t correctly spelled and user gets incorrect results
A: Google already uses this, but a spell correcting engine and a prediction engine should sort out both of them.
P: The products aren’t categorized correctly (the web shops tries to show the user more results so there are more categories then necessary)
A: I understand a web shop is trying to make money selling their stuff. But I’m looking for the most user-friendly option. I think that if you are rewarded with the most user-friendly web shop or something, people will come. So don’t cheat, be honest, and don’t sell you consumers crap!
P:
- You can only specify how you want to sort things using drop down menus
- When you sort things using the already annoying drop downs, you still get all the rest of the results (the act of specifying the most important feature of your purchase doesn’t eliminate part of the product range)
- You have to specify what range of price you use by sliders or by typing (which imho isn’t that user-friendly)
- Comparing results is very difficult and takes a lot of “forward / previous” browser action
A: These are four problems I would like to solve using one system. I want users to find their ultimate product quick and easy with the help of elimination.
Lets say the user gets their results fullscreen categorized in the following (so you’ll get three rows of products).
- Cheapest (most expensive is quite useless imho)
- Most bought
- Best rating (might also be “best value for money”, but should be something that has input from users)
Now, the user see’s how many results are in each category, and by choosing a category, he / she commits to wanting a certain property from the product that he / she will buy. After the user clicks a category, they’ll get the same screen. When the user sorts for prices, they get different pricing categories (0 – 50, 50 – 100, 100 and above, for instance).
When they do the same thing with the most-bought (& best rated) category they also get the price categories and see which one is rated the best in a certain price range (simply because if people are searching for the best rated, they’re not searching for a 3-4 star or a 1-2 star rating, they are searching for the best).
Now, after these rounds of elimination (the user ultimately clicks the category in which he / she thinks lies their perfect price / product), they might have as little as 4 results. Now we come to the part of comparing those four results.
I’m planning to use some sort of javascript / ajax pop-up system that doesn’t require the user to hit the previous button in their browser. Now, the comparing will also take place by elimination, if the user doesn’t want a product, he / she simply clicks an elimination button removing that product from the pile she has generated by searching. If al goes according to plan (and my mind has thought this trough properly) the user should end up with only one or two product to choose from (which would have required the consumer a maximum of 5 clicks ; “Search”, “Category 01″, “Pricerange 01″, “Elimination 01″, ”Elimination 02″)
P.s. Damn I’m productive in the train!
I literally learned about a majority of this, but however, I still believed it turned out valuable. Good task!
It was a good read, thanks for the share.