Ecommerce ops · Production

Algolia AI Search improves Plieger Groep's B2B e-commerce search, dropping average click position from over 16 to 5.5

The problem

Plieger's expert B2B customers — professionals who often knew exactly what they needed by item number or SKU — still could not find products due to poor filtering and search relevance. Both customers and employees complained, and employees were searching competitor websites to locate items before returning to buy from Plieger. The company also needed a solution that did not require a scarce and expensive dedicated developer to manage.

First attempt

Plieger's previous Solr-based search solution lacked the strong filtering capabilities essential for professional customers to find exact products by specification.

Workflow diagram · grounded in source
1
Professional customer initiates search
trigger
“They often know exactly what they're looking for, even the item number or SKU, but products can have diverse sizes, configurations, functionalities, or other characteristics.”
2
Autocomplete and query suggestions
ai_action
“it saw ACP drop from 12 to its current value over only 15 weeks after implementing autocomplete and search suggestions”
3
Dynamic Re-Ranking of results
ai_action
“Dynamic Re-Ranking, Facets & Filtering, Autocomplete, Recommend and Synonyms have had the greatest impact”
4
Facets & Filters narrow results
output
“Facets & Filtering has been essential, Heydendael says, and is "really making our clients happy." Now, installation professionals can easily narrow down products they need by specifications.”
5
Search analytics feedback loop
feedback_loop
“We've been able to become more data-driven. Before we didn't have any data, just complaints to work from. Now we can use the data from Algolia to really focus on the right KPIs”
Reported outcome

Over a 40-week period, Plieger's average click position dropped from more than 16 to close to 5.5, with ACP falling from 12 to that level in just 15 weeks after adding autocomplete and search suggestions.
The team became data-driven using Algolia search analytics instead of working from complaints alone.

Reported metrics
Average click position (ACP) over 40-week perioddrop from more than 16 to close to 5.5
Average click position (ACP) over 15 weeks after autocompleteACP drop from 12 to its current value over only 15 weeks
Reported stack
AlgoliaDynamic Re-RankingRulesSynonymsSearch APIRecommendVisual EditorFacets & FiltersQuery SuggestionsAutocompleteDynamic SynonymsSolr
Source
https://www.algolia.com/customers/plieger
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Over a 40-week period, Plieger's average click position dropped from more than 16 to close to 5.5, with ACP falling from 12 to that level in just 15 weeks after adding autocomplete and search suggestions.

What tools did this team use?

Algolia, Dynamic Re-Ranking, Rules, Synonyms, Search API, Recommend, Visual Editor, Facets & Filters, Query Suggestions, Autocomplete.

What results were reported?

Average click position (ACP) over 40-week period: drop from more than 16 to close to 5.5; Average click position (ACP) over 15 weeks after autocomplete: ACP drop from 12 to its current value over only 15 weeks (source-reported, not independently verified).

What failed first in this deployment?

Plieger's previous Solr-based search solution lacked the strong filtering capabilities essential for professional customers to find exact products by specification.

How is this ecommerce ops AI workflow structured?

Professional customer initiates search → Autocomplete and query suggestions → Dynamic Re-Ranking of results → Facets & Filters narrow results → Search analytics feedback loop.