Centering websites navigation around the search box - Good or Bad idea?

Question

I'm working on a website for a pharmacy store. The store would feature lot's of different drugs for different illnesses.

I had and idea to make 90% of the navigation center around the search bar. Meaning the website would have a normal navigation bar with sections for specific categories, but the landing page of the site would look basically like google.com with just a navbar and a big search box in the middle with a title "what are you looking for?" and a placeholder "Drug name or illness"

The user could input the name of a specific drug to look for it in the store catalog, or write in an illness or ache they are need a cure for.

For example writing "headache" in the searchbar would give the user an array of different painkillers suitable for headaches.

I was thinking if this is a good idea, and could actually help the websites ux.

Answer 1

It's difficult to say in black or white what is the best approach. I was working on a electrical component storage system in which I decided to try and do the "search first" approach. The primary issue I had was situations when the user wasn't sure what components they needed (ex. they wanted to browse by category). If they didn't know specifically what they were looking for they, couldn't find the object easily, or even at all. In other words, I made the system too search dependent for the task. However, in a pharmacy the user will already (hopefully) know what they want before they search, so you might not have to worry about that. If they don't know exactly what they want (as you mentioned "headache" in your post) it gets a bit more complicated.

The challenge is making a search system that works intuitively. Understanding "headache" is different from understanding "pain in my right leg" or "I need heart medication". I could assume that my users were competent enough to know "how" to search because it was in-house and I could teach them, but when people are presented with just a search bar you either need to make it really clear what they can search, or make sure whatever they do search gives relevant results. The former is probably a worse UX than browsing, and the latter is probably better (in my experience). However, building a smart search engine isn't a simple task.

Answer 2

There are some concerns with this approach.

First, if you intend for this to be the primary means of navigation, it would violate the UX principle of Recognition over recall. It is generally easier for users to find something if they do not have to know exactly what they are looking for, e.g., exact name, spelled correctly.

Secondly, to deliver usable results your tagging system would have to utilize a level of AI heretofore unknown for a search box. For example, if the user entered "heart" in the search box during February, expecting to see a list of results where she could choose a medication because she can't remember how to spell the one she needs, she'd get a long list of Valentine's day decorations. Search can't tell what she's looking for just by typing "heart." Or you'd have to segment results: Medications, seasonal, greeting cards, candy, nutraceuticals, etc. to prevent her from having to click through pages of results. Partial matches would be a nightmare too, considering overlap of brand names with concepts.

Thirdly, there are product categories that people would want to compare products in, and search result pages are typically not conducive to comparisons on details. You'd have to put so much product information on each search result that it would have to serve as a category page, but if the user accidentally hit the back button twice then she'd have to redo the search.

Your idea may be a nice feature if you can deliver quality content just by entering "heartburn," but maybe not a primary navigation system.