Web analysis: the key to successful e-merchandising (Part 1) Web analysis: the key to successful e-merchandising (Part 1)

Web analysis: the key to successful e-merchandising (Part 1)

by ,
on 18 November 2016


Several months ago, we reviewed basic e-merchandising techniques – highly recommended reading if you haven’t yet! We addressed the importance of acting on different levers: the tree diagram, search engines, product ranking in a page list, and facets but also the mechanics of cross/ad/up-selling and the need for testing to find the best combination for meeting your objectives and improving your results.

But even before you act on your e-merchandising, it is important that you ask the right questions in order to prioritize your actions.

Web analysis: the key to successful e-merchandising actions

Do you know how to determine the actions that will have the most impact on your business? In which lever will it be most profitable to invest first? Which optimization will require the least amount of effort with the most effect on your results?

I’m not reinventing the wheel here: if you want to analyze lever performance (and answer these questions), then above all you must first track and measure efficiently. Easier said than done in some cases. For example, if you have access to Google Analytics, note that the tab “Behavior > Site Search” identities traffic share using your search engine. But in reality, it does only half the work.

Allow me to explain: all visitors who use auto-complete (depending on the configuration) or yet again those who are directly redirected to a tree node further to a search are not included in this segment. So by looking at this data only, you run the risk of minimizing the search engine’s weight in your objective results.

Now imagine that you have this data and that it appears to represent 30% of your traffic from the home page. You’ll surmise that the quality of traffic using your search engine is low (little consulting of product pages, low transformation rate, weight in revenue generation – everything depends on the indicators you have decided to follow). So you’ll want to act and optimize this lever, and you’re right to do so. And so, you will look to understand the reasons for these results – is this due to a problem with understanding the engine, the relevance of the products suggested on the results pages, the facets (filters) available which do not really permit refining the results?

Note only that all these questions are likely to remain unanswered if you did not think to follow the “events” linked to each of your e-merchandising levers beforehand. In fact, the web analysis tools do not provide for native tracking and the data is not retroactive. You must wait at least a month before making a first analysis.

Event tracking: a miracle solution

From here, I can hear you saying “OK. I’ve got it! I promise to track everything from now on. But how?”

Nothing could be easier. Well, that is, if you have measurement experts at your side to help you draft a tagging plan that’s dedicated to your e-merchandising and also to implement it. No doubt, you’re used to tracking your pages but in the long run you don’t have any information about what happens on them… The solution to favor would seem to be to track by event – event tracking for GA – which will allow you to follow all your visitors’ interactions in a very precise manner.

1_web analyse

event actions – tracking and monitoring of facet use

Il serait peut-être judicieux de faire une capture d’écran en anglais

The idea is a simple one: 1 event = 1 action.

So it is important to list each action associated with your levers. Note that you can group all of them to one lever by using the notion of event category – which will make your life easier during analysis, I assure you. But to make things even clearer, let’s take a concrete example.

“As an e-merchandiser, I would like to understand the way in which visitors use my facets and their impact on my revenues in order to identify areas for optimization”: here, each facet: sex, color, size, fabric, etc., each associated value: XS, S, M, and L for the size facet, but also the position of the latter when they’re used will represent an event, in the “facets” events category. You will then be able to analyze the share of traffic using facets but also those clicked, the influence of their position in honing results and you will know the impact of each action on transformation. From there, you need only analyze the data and write your action plan to improve performance results!

To summarize

In a few words, here are the key steps to remember:

  1. Track the use of your e-merchandising levers as it’s not native
  2. Analyze the weight of each in reaching your goals
  3. Estimate the potential business of the actions undertaken
  4. Build your roadmap and prioritize, the famous effort/effect ratio
  5. Act by activating and measuring your optimizations

And don’t forget the approach: analyze, adjust, repeat!


, , , ,

, ,

to read next...

Leave a Reply

Your email address will not be published. Required fields are marked *


Web analysis: the key to successful e-merchandising (Part 1)