GRUBE Unlocks Real-Time Engagement and Boosts ROAS by 70% With Loomi AI

GRUBE adopted Bloomreach’s agentic personalization platform, Loomi AI, to understand and act on customer behavior — replacing static segments with a real-time picture of each customer’s unique wants and needs. 

Loomi AI serves as the intelligence layer behind every customer interaction, powering a dynamic, behavior-based interest scoring model that translates individual shopping signals into personalized email campaigns, triggered customer journeys, and precision-targeted print catalog distribution.

Individual interest profiles = a real-time understanding of each customer. Rather than assigning customers to broad segments, GRUBE leveraged Loomi AI to build individual interest profiles from a shopper’s behavior — searches, product clicks, purchases, campaign interactions, and more. Each customer receives interest scores per topic (e.g., hunting, forestry, outdoor apparel), updated continuously as behavior evolves. 

Behavior-triggered communication = engagement when interest is high. When a customer’s interest score in a specific category passes a defined threshold, Loomi AI automatically triggers the right campaign: a category-specific email, an on-site message, or a timely offer while interest is still active. GRUBE no longer waits for the next scheduled send to act on real-time signals.

Print catalog optimization = less print volume, more revenue. GRUBE used interest scores to redesign how its specialist print catalogs are distributed — only customers with sufficiently high, relevant scores receive a given edition, while low-interest recipients are excluded or receive a lighter format.

How It Works

1. Customer browses GRUBE’s site → searches, clicks, category views, and product interactions are captured as behavioral signals by Loomi AI.
2. Signals are weighted and scored → purchases and wishlist additions carry more weight than casual category visits; recent behavior scores higher than older interactions.
3. Interest scores update continuously → each customer interaction builds an evolving profile across topics (hunting, forestry, outdoor), reflecting current intent.
4. Score crosses a threshold → Loomi AI triggers targeted action: a personalized newsletter with category-relevant content blocks, a follow-up email, or an onsite message.
5. Print distribution is scored → before each catalog mailing, GRUBE filters recipients by interest score — only high-relevance customers receive specialist editions, reducing waste and protecting margins.
6. Results feed back into scoring → engagement, purchases, and opt-out signals refine the model continuously, improving precision over time.