Built a data-backed routing system that tripled high-value lead coverage using 3 months of real conversion data.
Website Form CVR
Highest-converting source — GOLD
Inbound Call CVR
Most valuable per-lead — GOLD
Chat CVR
13x lower than GOLD sources
Days to Close
Target after routing fix
Sierra's sales team had four agents handling inbound leads. All leads — regardless of source, value, or conversion likelihood — were distributed in whatever order they arrived. There was no routing logic, no tier system, and no awareness that a $5,000 website form enquiry and a cold chat message were worth treating differently.
Every agent had a category they converted well in — and categories they almost never closed. No one had mapped this before.
Six lead sources. Wildly different conversion rates. Zero differentiation in how they were handled.
Cart abandonment leads (751 per quarter) were consuming agent time at a 0.4% conversion rate. That's 5x worse than the next-worst source, and agents were spending real minutes on these. Meanwhile, inbound calls — 71.4% CVR — were sometimes going to whichever agent happened to be free.
Three-tier routing system
GOLD
Website Forms + Inbound Calls · CVR 63–71%
Route to Leslie first (highest close rate on high-AOV furniture), then Larry as backup. On Fridays: all GOLD leads go exclusively to Leslie — dubbed 'Leslie Day' in the routing doc. Target: 8-10 Gold leads per day for Leslie, up from 2-3.
SILVER
Product Inquiries · ~30% CVR
Route to Angela first (strongest on general and mixed inquiries), then Mollie. Standard 5-day contact cadence. Personalised outreach with category-specific pain angle.
BRONZE
Chat, Custom Forms, Cart Abandonment · <5.5% CVR
90% routed to automation. Mollie handles manual escalations. Bronze-to-Gold escalation path: if message contains 'bulk order', 'multiple items', 'designer', 'urgent', or order value >$5K — flag and route to Gold handler within 30 minutes.
Thursday was identified as the weakest day — only 2 of 4 agents available. Gold leads received on Thursday were held for a Friday Leslie callback, resulting in 25% better conversion than routing to a non-specialist agent same-day.
Week 1 (10% of leads): Route only Friday website leads to Leslie. Track conversion improvement. Gather agent feedback.
Weeks 2-3 (25% of leads): Add Thursday hold queue. Implement chat automation for Mollie. Include Larry in Gold routing.
Week 4 (100%): Full rollout. Activate all automation. Complete category routing. Calculate ROI.
Gold Lead Routing Accuracy
Random → 90%
Leslie's Daily Gold Leads
2-3 → 8-10
Chat Response Automation
0% → 90%
Avg Days to Close
5.2 → 3.5 target
What misrouting cost — per lead.
Each Gold lead misrouted to a non-specialist agent instead of Leslie represented ~$2,000 in lower expected conversion revenue at Sierra's AOV. Leslie's Gold target (8-10/day) vs. baseline (2-3/day) represented 5-7 additional Gold leads per day at $3,400 average order value.
Routing isn't sales support. It's revenue architecture.
The data existed for months. It just hadn't been organized into a routing decision. This was a 3-week analysis project that unlocked a structural revenue improvement. Most ops problems aren't engineering problems — they're data-reading problems.
Thursday's weakness was only visible after I built the weekly schedule matrix. Without that visual, the team would have continued treating Thursday like any other day. Always model the constraints before designing the solution.
Agent resistance was listed as a real risk. Showing each agent their own strong-category data — and framing routing as playing to their strengths, not limiting them — made adoption significantly easier.