Bloomio adds the BrowseIntent signal to Klaviyo as a custom profile property. Once the property is available, Composer, Claude, or ChatGPT connected through MCP can help you inspect it, find a starting point, and pick the first flow, campaign or segment to work on.
“According to Klaviyo, behavioral targeting ‘is definitely going to help with customer retention and optimizing for customer lifetime value (CLTV)’, and references a report from McKinsey that shows companies excelling at personalization generate 40% more revenue from related activities than average players.” — Klaviyo
Step 1: Confirm that the agent can read intent data
Ask the agent
Do you see a custom profile property named 'BrowseIntent'?
Step 2: Teach the agent about intent levels
Teach the agent what the property is, and what it means.
Ask the agent
Here is the definition of this property. What do you think and how can we use it in Klaviyo?
We define browsing intent as a practical way to estimate shopper intent from recent browsing behavior. The model produces an intent score, which is then grouped into high, medium, or low intent levels. The intent score is inferred from behavior, not observed directly. The model is heuristic and intentionally simple. It evaluates six behavioral signals over a moving 90-day window and ranks each eligible profile relative to other active profiles for that merchant. The data model combines weighted behavioral signals with relative audience ranking, so each eligible profile is evaluated in the context of the merchant's own traffic patterns rather than against a universal benchmark: 1. Recency 2. Frequency 3. Cumulative session depth 4. Pageview or product-view activity 5. Add-to-cart activity 6. Checkout-started activity.
Step 3: Discover where to focus first
Walk through a brief discovery process to define one specific test or task with a clear outcome.
Ask the agent
I want to find the best opportunity to use BrowseIntent in my Klaviyo account. Ask me questions one at a time to help me figure out where to focus first.
More examples
1. Build the 3 intent-tier segments
This is the foundation for everything else. Once built, you'll have audiences you can send to, analyze, and sync to ad platforms.
Ask the agent
Create segments for BrowseIntent = high, BrowseIntent = medium, and BrowseIntent = low.
2. Analyze existing campaigns against intent-tier audiences
Once the segments exist, ask:
Ask the agent
Compare open rate, click rate, and revenue per recipient across my BrowseIntent high, medium, and low segments for my last 5 campaigns.
3. Audit flows for intent-based personalization gaps
Ask the agent
Audit my abandoned cart and browse abandonment flows and tell me where I could add a conditional split based on BrowseIntent.
4. Draft a campaign targeted at high-intent browsers
Put the segment to work right away.
Ask the agent
Create a campaign targeting my BrowseIntent = high segment with a conversion-focused offer.
5. Build a browse abandonment flow that branches on intent
The most strategic use of the property.
Ask the agent
Build a browse abandonment flow that splits on BrowseIntent — high intent gets a fast, urgent SMS and email sequence, medium gets a standard nurture, low gets a single brand reminder email.
Review before anything is activated
Start with read-only discovery, then ask the agent to draft the audience, message, or flow. Review the profile property, audience rules, send conditions, channel eligibility, and message before activating anything. The agent can help prepare the work, but the final decision stays with you.
What to measure first
Pick one outcome and compare the intent tiers against a baseline or holdout when you can. Use placed-order rate and revenue per recipient as primary measures. Open and click rates can help diagnose the message, while unsubscribes and SMS opt-outs show whether the channel and intensity are appropriate.
According to Klaviyo and McKinsey, behavioral targeting and personalization drives 40% more revenue.[1]
[1] McKinsey: The value of getting personalization right—or wrong—is multiplying
Category
Lifecycle marketingTags