Improve your CRO thanks to good results ranking on your e-commerce

Sam Smith

Ecommerce Consultant at Doofinder

• 
• 
• 

We are all different, we each have our tastes, our preferences, our habits. It’s the same case for the users who arrive on your e-commerce: they are not looking for the same thing, nor in the same way.

When buying online, many factors influence the decision-making: the trust placed in the e-commerce, the price-quality ratio, the reputation of the brand, the content, the means of payment. As an e-merchant, we must be attentive to these factors in order to increase our conversion rate.

The post is based on feedback from e-merchants clients. We are going to see that the results page ranking has a positive and quick impact on KPI’s as: rebounce rate, number of pages viewed, number of orders… And that it can even affect the website SEO, via positioning keywords relevant for the e-commerce.

Let’s go!

1. Analysis of the pages to optimise

You’ve got it, today we are speaking about the ranking of results page or categories in e-commerce. The first step is to analyse the conversion rate for each of the categories and results page. From this starting point, we will be able to set objectives and identify the strengths and weaknesses of our website.

If you use Google Analytics and have the option e-commerce on, you just have to select the organic traffic segment and start the analysis like this:

Behaviour > Site content > Landings Pages

You will see the conversion rate for each page of your online shop.

Image: Analyse of the conversion rate of the landings (Source: Google Analytics).Select a time interval (we advise more than one month) and export the excel file. Then, filter the categories. The categories that can be optimised are those with a number of sessions and a conversion rate lower than the average.

2. What factors should I consider before ranking the lists?

The sector and the characteristics of your website, the seasonality and the kind of users are fundamental aspects to consider when ranking the results page. You can also choose if the ranking will be permanent or temporary.
Product and messages
Which products do you sell? What are the values you want to communicate on? What advantages do you have over your competitors?
Your ranking strategy will differ according to your positioning and the messages you want to communicate. For instance, you will not have the same strategy if you sell cheap shoes or if you focus on selling the latest brands at a medium-high price.
Seasons and special events
During sales, Black Friday or for special events such as Valentine’s Day, Mother’s or Father’s Day, the user changes their search habits as well as the typology of the products that are searched. So it’s important to take in consideration the seasons and the special events when doing the ranking, so that the results returned correspond to the expectations of the user.
Typology of users
It is an important criteria to take into account when classifying lists. As it is difficult to personalise the ranking according to each profile or user, we are going to do it by segment. Your users are divided into different groups (see typology below) according to their interests and priorities. Knowing the main groups will allow you to choose 2 or 3 criteria that will include the most part of the users.
Here are the most frequent users typologies:
* Look for the cheapest products
* Look for the newest products, no matter what the price
* Look for the best percentage of discount
* Look for the most famous or top rated products
* Look for the best sellers
* Look for a mix of all these criteria

3. Recommandations in order to optimise the ranking

Some of these recommendations need to modify the code temporarily or permanently. You might need the help of your technical team.

Basic configurations

The users of your online-shop should be able to classify the products according to the following options:

• Price (from the cheapest to the most expensive).
• Discount (from the biggest rate of discount to the lowest).
• Newest Arrivals (from the newest to the oldest)
• Best Sellers (from the most sold to the less sold)
• Top Rated (from the highest numbers of positive reviews to the lowest)

It seems easy, but many e-commerce companies do not offer theses possibilities of classification. Yet, they allow the user to choose quickly the way they want to classify the products.

These options of classification are enough for most of the users. However, a new group of users may want to classify the products according to a different order. We shouldn’t neglect any possibility: the criteria can evolve over time.

Visibility of the option “Sort by”

In most of the e-commerce sites we are working, this functionality, as well as the functionality “Filters” is heavily used by the user when he arrives on the results page. So it’s important to make it visible, as well as easy to use and integrated in the mobile version of the website.

Advanced manual ranking

With the basic configuration, each user can choose to rank the products according to their favourite criteria, however… which order should we show by default?

When a user lands to your website in an organic way, after a google search for example, the page on which they arrive has a default order. Nothing can be left to chance and it’s important to choose the order by default. For that, we are going to use the logic and the criteria we saw previously.

I’m going to explain the way I create the “Personalised ranking”, above all for clients with a reduced catalogue and low product rotation. It involves manual work and you won’t necessarily have the time to do it on a daily basis, in spite of the improvements it can produce.

We are going to use an excel file and follow the following steps:

1. Choose the category for which we want to rank the products (according to the criteria we’ve seen above).
2. Download all the products included in the category, as well as the data about them: product ID, SKU (stock-keeping unit), title and price.
3. Add the columns “number of sales”, “number of client reviews”, “average review rating” (if the rating is 3 or less, it’s better to put -1, -2, etc.), “most recent” and “discount”. Note: you can choose the laps of time you want to select this data, but we advice to choose one year.
4. Fill the data for each column and product. You can find the       number of sales within Analytics and the rest is available in the     category itself.
5. Once the spreadsheet is filled, we are going to apply a formula with multiplying factors for each field. These coefficients are determined according to the weight of each criteria in the decision to purchase.

Example of a formula with multiplying factors: (Sales * 0,4) + ((Number of reviews * Average of the reviews rating) * 0,3) + (Discount without percent * 0,05)
Example applying to a product: A mattress was sold 10 times during the year, it has 5 reviews with an average rating of 4,4 and a 10% discount. The formula applied to the mattress will be: (10 * 0,4) + ((5 * 4,4) * 0,3) + (10 * 0,05). The final rating is 11,1.
6. Rank the product ratings in descending order and add another column for the ranking of the products (1, 2, 3, etc.).
7. We have not forgotten the most recent nor the price. To do this I suggest inserting new products and the cheapest ones in the prior list. For example: every two products, add a new product, then every two products, add a cheap product.

Congratulations, you’ve created your by default list! All that is left to do is apply this manual ranking to the back office of the shop.

You can of course modify this formula by changing multiplying factors according to the weight you want to give to each criteria.

Automatic advanced ranking

For the clients who have a very large catalogue or a catalogue which often changes, ECOMMBITS develops functionalities allowing the quick application of this mix of criteria, directly from your back office.

This formula is not as accurate as the one we’ve just seen (manual ranking) but it allows it to be very reactive and to quickly test the changes produced.

4. Results

Improvement of the conversion rate

Let’s take a look at a real case of a client from the textile sector and the comparison of its conversion datas in 3 optimised result pages. 5 months before the optimisation and 5 months after.

We see that the conversion percent increases from 0,48% to 0,56% for the average of the categories. In other words, the conversion rate increases by 16,6%. If the months before the optimisation of the conversion rate would have been the same, KPI’s like the number of orders or the income would have been respectively 17,1% and 16,6% higher.

Improvement of the usage datas

Besides the conversion rate, for which we have a specific interest because it has a direct impact on the business, other KPI’s also improve. The time spent on page, the number of pages visited and above all the rebound rate are significantly improved. Indeed, now, the user finds the products they are looking for faster and they keep surfing on the website until they buy.

Improvement of the positioning

Finally let us take a look at an indirect consequence of the improvement of the conversion rate and the using of data. This is the improvement of the positioning for the keywords which led to these pages from Google.

For Google, it is fundamental that the landing page correctly answers the search of the user. As this has been improved, Google rewards it by improving the positioning for this landing and the related landings.

The following screenshots show the main page before and after the improvement, as well as the SEO of its main keywords.

The landing has been improved for the 3 main keywords and the traffic received by the landing increases from 737 views to 1.100 views, that is a 49,25% rise.

Better efficiency of the marketing actions

Thanks to these ranking criteria, we create a specific url. It works for the users who arrive on your online shop through organic traffic and begin to search for products. Imagine now that you were bringing users directly onto these pages users thanks to paying campaingns like AdWords, Facebook Ads, etc.

The impact is generally seen within a short period of time: the conversion rate increases and one of the KPI’s which most interests us in marketing, the CPA (Cost Per Acquisition) decreases significantly.

Result: your marketing plan ROI is multiplied.

5. Conclusions

To adapt our e-commerce to the user by allowing them to sort the results page as they want is already a great improvement in terms of UX. However, as we have seen, to modify the way products are ranked by default in a category usually gives very positive results in terms of conversion rate, improvement in the usage of data and even improvement of the SEO campaigns.

To attract traffic is very important, but without a good user experience, the marketing implemented won’t be very useful or, at least, won’t be as profitable as it could be.

Little changes as we’ve seen allow you to improve your relationship with the user. However, the personalisation or the adaptation of the content according to the interests of the user goes much further (banners, messages, discounts, etc.).

Are you ready to try?