01 – Tackling the Information Architecture
Understanding shopping behaviours on RedMart
In order to understand the reasons behind why users are choosing (or not choosing) RedMart over its competitors, as well as understanding how user shop for the things they want or need, we conducted user interview with 7 RedMart users. We asked questions like what do they look out for when they shop online, what are important to them etc.
Also, we tasked them to think out loud while browsing through the site, and observed their behaviours when browsing through categories when they look for certain items without using the search function.
Here’s the summary of our findings:
- Most users already have a list of things they want to buy, hence search is a go-to function for them;
- Navigation Menu is unorganised and confusing, finding it frustrating to go through a long list of categories;
- Users are unsure of certain category names and what’s inside the categories exactly (eg Stone fruits, Large Fruits);
- Users are visually driven when they shop online;
(Users who prefer to buy groceries in-store itself because they can view everything at a glance, pick and choose on their own and ensure freshness of items (for Fruits and Vegetables) and not near expiring dates)
- Users hardly make use of filter options.
With a general understanding of how users shop online, we proceeded with understanding how we can help to enhance the shopping experience when they search for things in different categories. With the tight timeline we had on hand, we focused on 2 main categories – Fruit & Vegetable and Baby
Fruit & Vegetable Category
We first conducted a quick user interview to have a general understanding of how users would shop for Fresh Fruit & Vegetable. All users mentioned that freshness is an important indicator.
The idea was to strip away RedMart’s Fruit & Vegetable current categorisation and allow participants to categorise them in a manner they desire and coming up with new categories, should there be any. This also allows us to compare against RedMart’s current categorisation and identify the differences, should there be any as well.
Open Card sort for FRUITS
32 participants, 37 fruits
Open Card sort for VEGETABLES
open, 32 participants, 45 vegetables
We analysed the results from the Dendograms using Actual Agreement method. Both results for Fruit & Vegetable showed that the way participants group Fruit & Vegetable were were largely similar to current RedMart’s categorisation. However, there were a few lonely items that were inconsistent and had very low percentage of agreement. Kiwis, peaches, figs, bean sprouts and zucchini to name a few.
Hence we decided to keep RedMart’s current categorisation and tackle those lonely items. In order to verify if our participants indeed have problem finding those lonely items on RedMart’s current site, we conducted a Tree Test with a total of 12 tasks.
Red indicates failure of task and
Green indicates success of task.
The Tree Test results shows that participants indeed have problems finding those lonely items on RedMart’s current site IA.
After analysing and synthesising the card sort results, we identified the best match for the lonely items and gave them a new category.
Also, as mentioned, users are unsure of certain category names. Ambiguous naming such as Tropical Fruit and Exotic Fruit confuses users as it requires them to decipher and think. Therefore, we simplified and combined the category names that users had trouble with.
Putting everything we had together, we came up with a revised IA for Fruit & Vegetable.
Baby, Kids & Toys Category
As the entire category is too huge to tackle at one go, my team decided to focus on just Baby.
Before we apply the same methodologies we used for Fruit & Vegetable, we first conducted a user interview with experienced mothers to better understand their buying behaviours as none of us are mothers ourselves, thus lacking the knowledge. We wanted to find our what are the top essentials they purchase for their babies (or for themselves even), and things they look out for, do they have any brand loyalty etc.
We interviewed a total of 7 mothers and here’s a summary of our findings:
- Mothers always have a list of to-buy items;
- Mothers are price sensitive, they make purchases with the highest cost savings;
- Mothers purchase online solely due to convenience – things get delivered to their door step; and
- Most commonly purchased essentials for baby items are: Baby Wipes, Diapers, Milk & Powder, Toiletries.
With that in mind, we proceeded with Hybrid Card Sort.
1st Hybrid Card sort for Baby
6 participants, 12 categories, 77 cards
The first card sort was to determine if users would categorise the items similar to RedMart’s current site. Our results showed that mothers were categorising their items differently and inconsistently as compared to RedMart’s. Most users sorted their cards accordingly to the needs for mothers and needs for their babies, or they would group items to places where they belong, for example, things from the bathroom, or things meant for outdoor.
We concluded that RedMart’s current categorisation for Baby can be improved on. Therefore, we ran a second Hybrid Card Sort.
2nd Hybrid Card sort for Baby
12 participants, 5 categories, 77 cards
We had 5 main categories (derived from top essentials from our interview), and left the our users to create new ones. We wanted to discover how mothers group the other products to determine the remaining categories.
As we had lesser participants this time round, we made use of the similarity matrix to analyse our results. Our results also showed that there wasn’t much inconsistency for our second Card Sort. Therefore, we put together what we had and came up with a revised IA for the Baby category.
With the new and improved IA for the categories, we moved on to tackle the second part of the brief – improving the navigation design.