Ecommerce ops · Production

Bike Totaal drives in-store conversions with Algolia AI Search

The problem

Dynamo Retail Group's Elasticsearch-based search delivered slow, irrelevant results — a search for a 'black bike' surfaced a black bag first — and required external developers for even small configuration changes, while the customer base was shifting to more digitally-savvy users.

First attempt

The existing Elasticsearch-based search engine produced irrelevant results and required external developers for every small change, making it impossible for the ecommerce team to iterate independently.

Workflow diagram · grounded in source
1
Product data sync via Alumio
integration
“Data is pushed to Algolia via APIs from Alumio's low-code integration platform”
2
Affiliate store index in Algolia
integration
“Dynamo created an index of its affiliates in Algolia. "If a customer wants to book a test ride, they can select the store — and all of it happens through Algolia"”
3
Customer initiates search
trigger
“People who use the search bar on a category page are more likely to convert than those that don't”
4
AI search with Dynamic Re-Ranking
ai_action
“Algolia Search, Dynamic Re-ranking, Categories, Rules, Merchandising Studio, Analytics, Dynamic Synonyms and A/B Testing”
5
Unknown query routing
routing
“we noticed a rise in search queries containing 'fat bike.' We don't sell them, but now at least we can redirect customers to a page that shows them alternatives”
6
Results rendered in Storyblok CMS
output
“retrieved and rendered in Storyblok's headless CMS, giving Dynamo full control over indexing and display”
7
Merchandiser configures rules
human_review
“We can immediately see what we're doing — an exact copy of our category page — and how it responds when we add rules or push products. It's a nice feature. It makes things easier because you see it live”
8
Analytics and A/B testing feedback
feedback_loop
“Analytics, Dynamic Synonyms and A/B Testing”
Reported outcome

After switching to Algolia, Dynamo saw increases in search usage, click-through rates, and conversions from search, received overwhelmingly positive feedback from customers and store owners, and can now manage all search and discovery in-house without external intervention.

Reported metrics
Search usageAn increase in search use
Search result accuracyMore accurate search results
Click-through rateGreater click-through rates
Conversions from searchHigher conversions from search
Show all 6 reported metrics
search usageAn increase in search use
search result accuracyMore accurate search results
click-through rateGreater click-through rates
conversions from searchHigher conversions from search
customer and franchisee feedbackoverwhelmingly positive
team self-sufficiency for search managementmanage all the aspects of search and discovery itself, without outside intervention
Reported stack
Algolia SearchDynamic Re-RankingMerchandising StudioDynamic SynonymsA/B TestingAlumioStoryblokElasticsearch
Source
https://www.algolia.com/customers/bike-totaal
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

After switching to Algolia, Dynamo saw increases in search usage, click-through rates, and conversions from search, received overwhelmingly positive feedback from customers and store owners, and can now manage all sea…

What tools did this team use?

Algolia Search, Dynamic Re-Ranking, Merchandising Studio, Dynamic Synonyms, A/B Testing, Alumio, Storyblok, Elasticsearch.

What results were reported?

Search usage: An increase in search use; Search result accuracy: More accurate search results; Click-through rate: Greater click-through rates; Conversions from search: Higher conversions from search (source-reported, not independently verified).

What failed first in this deployment?

The existing Elasticsearch-based search engine produced irrelevant results and required external developers for every small change, making it impossible for the ecommerce team to iterate independently.

How is this ecommerce ops AI workflow structured?

Product data sync via Alumio → Affiliate store index in Algolia → Customer initiates search → AI search with Dynamic Re-Ranking → Unknown query routing → Results rendered in Storyblok CMS → Merchandiser configures rules → Analytics and A/B testing feedback.

Related ecommerce ops cases