I am an analyst and these are the 4 lessons I have learnt from Black Friday (and the actions to get ready for the next Sale)

As the Christmas campaign rapidly approaches and new challenges come along with it, I wanted to spare a moment to reflect on what I have found interesting and teaching from this last Black Friday. First, let me explain why these learnings and derived actions are of key importance:

  1. Black Friday is one of the most important days of the year for eCommerce. For some sites it is even the day in which the highest gross revenue of the year is made.
  2. The experience on the site is almost as important as the product itself. If the user has a bad experience during their visit, it is most likely that they will not buy and will never come back.

The learnings and actions mentioned on this post are about Black Friday, but they are applicable to any Sale and other big promotions. The focus is onsite Search, but remember that the learnings from Search are not only used to improve Search.

1. Black Friday started early and had a higher CVR than the actual day

The problem:

The promotions on Black Friday are very attractive for the consumer, which is the main reason why so many people make the decision to buy during this day. The problem is that there is such an amount of promotional messages running on sites and in our inbox, that the consumer feels saturated.

I’ve recently been to Tokyo, which impressed me, among other things, by the high number of neon signs on their buildings. There are so many that in the end you don’t really pay attention to any of them, in the end you just see colourful lights. Internet becomes an online Tokyo on Black Friday and many of the promotional messages go totally unnoticed.

The solution:

What we have seen thanks to the data, is that those that start Black Friday a few hours (even 1 day) before the actual day, present a higher CVR. In most cases, that initial hour of the promotion has the best CVR of all Black Friday hours when it takes place before the Friday.

The traffic is not as high as it is during the busiest hours on Friday, but it is still good.

2. Higher zero results rate

The problem:

It is very frustrating to go into a website, search for something and get no results. No even 1 result, just a sad page showing that there is nothing for you. Sometimes the page even apologisies, which is nice, but not enough.

On Back Friday the number of queries with no results increases and the reason is that there is a longer long tail with a bigger variety of searched terms, which some of them are less generic than in a non-promotional period. The natural effect that these new terms have in the search engine, if the engine is not prepared for it, is an increase of the number of queries with no results.

The solutions:

  • Spellcheck — Increase number of misspelled letters:

When Spellcheck is activated, the search engine will look for similar terms when a term is misspelled. If I want to search for ‘sweatshirts’ and type ‘swetshirts’ instead, I will get sweatshirts as results even if my query wasn’t correctly written. Spellcheck’s set up requires selecting what number of misspelled letters are allowed. The higher the allowed number, the more the suggested terms that the search will provide and therefore the more results that the search engine will find.

  • Description field relaunch:

Some sites prefer to not show products in the results page that only contain the searched term within the description field in the product page. If I search for ‘Dress with stars’, and the query is searched in the description field, I could get products as results that are described as “Wearing this dress you will feel like a star”, and not because there are stars on the dress. To avoid generating noise in every search, we sometimes recommend to relaunch the query against the description field when the query has no results by searching in other fields (like name, category, etc…). The data shows that by doing this, the percentage of queries with zero results decreases.

  • Contextualise with Spellcheck and Description relaunch:

Contextualise shows the most relevant products for the user first in the search results page. The most relevant results are calculated based on the user’s affinities, which are created from the information of what content and what products they browse and buy. Adding Contextualise to any of the 2 actions mentioned above will not only reduce zero results, it will also show at the top the preferred products by the user, which will improve the performance of the query and the overall experience. The data visualisation above this point shows the growth in Findability and CTR when a query is contextualised vs when it is not.

3. Black Friday is the searched term with highest number of no results. Same as Sale

The problem:

Speaking of no results, both Black Friday and Sale, present a high number of searches during promotional periods. Users search exactly for the terms Black Friday and for the term Sale. These are very generic searches, which in many cases don’t provide any result. The solutions are very simple but still many sites don’t do anything about it.

The solution:

  • One possible solution is to have a redirect when these terms are searched and send the user to the Sale landing page, or to a product list page showing the discounted products.
  • Another possible solution is to add a new field in the product information containing if it is included in the Black Friday promotion or in the Sale. In this case, the search results page will show all the products that have Black Friday or Sale in their information.

4. Increase of searches by Reference

The problem:

One of the things that surprised me when I started to analyse onsite search data is the amount of queries that contain SKU numbers. They could be around 5% of the total searches, which is not little. What is even more surprising is the very high percentage of queries that these SKU numbers represent when looking at queries with no results. The percentage could be 25% or even higher.

The solution:

  • Create an autocomplete for SKU numbers like the one we created for Zara. The autocomplete shows the user what is the structure of the SKU number
  • In addition to that is good to show related products to the product that was searched by its SKU number if it is out of stock

At EmpathyBroker we wanted to share what the data and the experience have taught us about Black Friday, and what is even more important, what actions to take to improve the performance of such an important day. As the Chinese proverb says “Learning is a treasure that will follow its owner everywhere”, however, if no actions are taken from what was learnt, no improvement will happen.

Contact us if you want to learn more about optimising Black Friday and any other Sale. We make things happen :-)

Like what you read? Give Barbara Mackey a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.