How better search & personalization helped PC Componentes grow search click-through and conversion rate

PcComponentes was founded in 2005, by two entrepreneurs, and an initial investment of €5,000. The company started with a very clear goal — to offer end customers a unique purchase experience for electronics, laptops, mobile phones and home appliances. 

Currently, they have more than 30,000 SKUs, 4,000 physical points of order delivery all over Spain, as well as stores in Madrid and Murcia. 

From its inception, PcComponentes has worked hard to build much more than a brand for its customers. The goal has always been to build a team in which customers can trust, and establish a long term relationship with the company. In order to build trust, PcComponentes has always focused on listening to customers, understanding their requirements, knowing their preferences, offering what they need, and rewarding their loyalty. 

PcComponentes closed 2018 with €360m in sales, which represented 15% growth year over year. Additionally, 1.5 million orders were placed, with 5.6 million products sold. Ten percent of all annual bookings took place on Black Friday, in just under one week.

The challenge

PcComponentes’ search experience wasn’t keeping pace with their fast growth

One of the main challenges was ensuring fast performance of PcComponentes’ search engine. During peaks such as Black Friday, where PcComponentes’ users expect the best possible experience, price and offer, the website reaches order volumes of 1 order per second, and the search engine simply cannot slow down or underperform. 

In addition, PcComponentes’ product catalog is growing at an exponential rate and keeping it synchronized with the search index was a challenge. With more than 30,000 SKUs that are continually being updated from different systems, maintaining, updating and sending that product detail information to the search engine was not an easy task. 

In search of its new search platform, PcComponentes also wanted to ensure high availability and low latency to provide its users with a great online experience that delivered fast, relevant results. Finally, ease of implementation and the ability to scale with a growing user base were key characteristics PcComponentes was looking for in the solution.

The solution

Algolia Search-as-a-Service Platform, Analytics Dashboard, and Algolia Rules

After its evaluation of different build and buy options, PcComponentes selected Algolia. 

PcComponentes leverages Algolia’s Rules to tailor search for its business needs and adjust specific queries to increase the click-through rate for those queries. In addition, PcComponentes uses Algolia’s analytics dashboard to detect zero result queries - what users are searching for, but what may not currently be in its product catalog - to continue to optimize the search and discovery experience. 

In terms of performance, PcComponentes uses Algolia’s analytics to measure the click-through rate and the conversion rate by connecting and joining with data in its other systems. 

Finally, adapting any external system to a core system is always a challenge. PcComponentes partnered closely with Algolia’s customer success and technical team to effectively integrate Algolia’s search API with PcComponentes core system. PcComponentes was able to leverage Algolia's set of backend API clients to build a custom integration for its unique needs with the Algolia team. 

The result

PcComponentes nearly doubles click-through rate and quadruples conversion of website search visitors versus the website average

With Algolia, PcComponentes was able to quickly and efficiently index a product catalog with over 30,000 SKUs. Additionally, they were able to increase the click-through rate of search results by 81%, through the use of analytics and personalization. 

Search users on the PcComponentes convert at a rate 4x higher than the website average. Performance requirements were readily met with Algolia with an average 78 millisecond response time for search queries, and 100% availability in the trailing 90 days.

What's next 

Voice search is a concept and approach that is entering our daily life, and PcComponentes has seen the trend of end users making use of voice search and messaging more and more in everyday life, avoiding typing or texting altogether. In the case of e-commerce, PcComponentes sees this trend to follow exactly the same. 

One of the requirements is to be able to interpret the nuances of each spoken language. With Algolia, there are already voice and NLU (natural language understanding) modules that PcComponentes will begin implementing soon.