Petit Bateau optimizes its internal search engine transformation by 56%! Petit Bateau optimizes its internal search engine transformation by 56%!

Petit Bateau optimizes its internal search engine transformation by 56%!

by ,
on 12 April 2018



“Aware of the lost revenue generated by our search engine, we wanted to get a handle on this lever in order to guarantee our customers an optimal search experience, and all the more so, to improve performance.”
Karine Attali, Online Sales & Web Merchandising Manager with Petit Bateau

  • improve the search engine relevancefor strategic requests and international long train (six languages) > reduce the exit rate for search engine users, increase transformation for visitors using the search engine and reduce the number of searches generating 0 results.
  • offer a real search experience by going beyond a product suggestion > rapid access to categories, highlighting boutiques, increasing visibility of brand-created content
  • enable the e-merchandising team autonomy in its optimization actions >reactivity on the merchandising strategies to implement depending on promotional operations, AB testing on the strategies, contextualization of the filters to display, etc.


The project’s major challenge was first and foremost to re-empower the e-merchandising team in its optimization actions. To do so, it was necessary to have available a merchandizing-oriented tool and not a technical one. The decision was made to go with the Prediggo solution. It is true that in addition to being “user friendly”, this tool gives e-merchandisers the opportunity to configure their levers with fine tuning, based on their objectives.

Once the tool was chosen and the parameters set, we concerned ourselves with the manner in which to add value to the Petit Bateau offer (by suggesting boutiques but also content) and to facilitate access to list and product pages: the objective being to maximize the penetration rate for categories and thus, product visibility. This translated to enriching auto-completion which previously suggested product names only.

We then worked on the choice of product attributes to index in searches in order to guarantee the relevance of results and in so doing, reduce the number of requests with 0 results. Note: the more you index, the greater the risk of creating “noise” in the suggested results which deteriorates relevance!

To finish, we created several ranking strategies, one per important promotional event, to meet the objectives within each.

Extra information: in order to make the search bar more visible (and while waiting for the header redesign) the latter went from blue to white

The search engine in november 2016


Period of analysis: October 2017 vs October 2016 – average across all countries

  • As a result of the optimization of the textual relevance (thought given to attribute indexing in the search and creation of synonyms), the rate of requests leading to 0 results decreased by 19% (below 1% for France).
  • As a result of improving the relevance of the products proposed, the conversion rate for visitors having searched increased by 56% and sales revenue increased by 70%.
  • The drop in the exit rate after searching, while low (-2%) also shows improvement in the overall relevance of the search engine.

The search engine in november 2017

So. Have we convinced you of the necessity of offering your visitors a better search experience?



, , , ,


to read next...

Leave a Reply

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


Petit Bateau optimizes its internal search engine transformation by 56%!