00 / CASE STUDY

LMTY: Homepage Variant Testing

Client: LMTY (AI competitive intelligence platform)
Test Date: March 2026
Methodology: Synthetic user testing, 40 runs, 10 B2B SaaS personas

Executive Summary

Tested two LMTY homepage variants against 10 B2B SaaS personas (PMM ICs, Sales Leaders, Founders, RevOps) to measure conversion likelihood.

Key Finding: Neither variant is universally superior. V2 optimizes for C-level/strategic buyers (+34 pts CRO, +35 pts PM) but alienates tactical ICs and burned buyers (-41 pts VP Mktg, -13.5 pts PMM IC).

Recommendation: Segmented messaging strategy—serve V2 to enterprise/exec traffic, V1 to PMM/mid-market.

40
TOTAL RUNS
10
PERSONAS
2
VARIANTS
2
DAYS
BRIEF TO REPORT
01 / OVERALL RESULTS

CONVERSION RATES

VARIANT
AVG CONV.
95% CI
VS KLUE
SIGNIFICANCE
Variant F+ (V1)
50%
37%–63%
+2 pts
Not significant
Variant F+ V2
51.4%
38%–64%
+3.4 pts
Not significant
Klue (baseline)
48%
Baseline
AlphaSense
43%
-5 pts
Weaker
Crayon
28%
-20 pts
Weakest

Statistical Significance

V1 vs V2 difference (+1.4 pts) is NOT statistically significant (overlapping confidence intervals). However, persona-level differences are dramatic and meaningful for segmentation strategy.

02 / PERSONA-LEVEL RESULTS

CONVERSION BY PERSONA

PERSONA
V1 AVG
V2 AVG
CHANGE
INTERPRETATION
PM (David)
50%
85%
+35 pts ⭐⭐
V2 clarity wins
CRO w/ Klue (Elena)
32.5%
66.5%
+34 pts ⭐⭐
Value chain fix critical
RevOps (Amanda)
25%
56.5%
+31.5 pts ⭐⭐
Pricing + value chain
VP Sales (Marcus)
63.5%
63.5%
0 pts
No change
Seed Founder (Priya)
36.5%
48.5%
+12 pts ⭐
CI acronym fix
Series A Founder (Raj)
25%
25%
0 pts
Still skeptical
PMM Dir w/ Klue (James)
31.5%
27.5%
-4 pts
Switching costs
PMM IC (Sarah)
85%
71.5%
-13.5 pts ⚠️
V2 abstraction hurt
Sales Ops (Lisa)
55%
37.5%
-17.5 pts ⚠️
Lost specificity
VP Mktg w/ Crayon (Tom)
73.5%
32.5%
-41 pts ❌❌
"Agents" = broken promises

Key Pattern

V2 optimizes for C-level/strategic buyers (+34 pts average for CRO, PM, RevOps) but alienates tactical ICs and burned buyers (-24 pts average for PMM IC, Sales Ops, VP Mktg). This suggests segmented messaging strategy rather than universal homepage.

03 / KEY CHANGES V1 → V2

VARIANT DIFFERENCES

✅ FIX

1. Spelled Out "Competitive Intelligence"

V1: "Hire AI CI agents"
V2: "Hire AI competitive intelligence agents"

Impact: +12 pts for Seed Founders (eliminated CI/Continuous Integration confusion)

"Thought 'CI' meant Continuous Integration (software testing), not Competitive Intelligence. Evaluated LMTY as GitHub Actions competitor!"

— Seed Founder (Priya), V1 Run 2
✅ FIX

2. Made Value Chain Explicit

V1: Vague Sales benefit
V2: "PMM uses LMTY → Sales gets intel in Slack → Win rates increase 22%"

Impact: +34 pts CRO, +31.5 pts RevOps

"She immediately sees: PMM uses tool → Reps get intel in Slack → Win rates up 22%. No ambiguity about who does what or where value comes from."

— CRO (Elena), V2 Run 1
✅ FIX

3. Clarified Pricing Model

V1: "$299/mo Pro"
V2: "$299/mo per team (not per seat)"

Impact: Removed major objection for enterprise buyers

"$299/mo per seat or per team? 12 reps = $43K/yr vs $3.6K/yr = massive budget difference. This is a blocker."

— VP Sales (Marcus), V1 Run 2
⚠️ NEUTRAL

4. Reframed ROI for Founders

V1: "Saves PMM 8hrs/week = $16K annual value"
V2: "Enable 2 launches/quarter → 6+ launches (3x output)"

Impact: None (founders still skeptical of both framings)

"'2→6 launches' sounds like SaaS marketing hyperbole. Show me the receipts."

— Series A Founder (Raj), V2 Run 1
❌ BACKFIRE

5. Added "Agents = Employees" Metaphor

V1: "Unlike Klue/Crayon (you filter/post), LMTY automates"
V2: "LMTY agents are employees, not tools. They monitor, filter, synthesize, post automatically."

Impact: -13.5 pts PMM IC, -41 pts VP Mktg

"After Crayon failure, 'agents automate everything' = empty promise without specifics."

— VP Mktg (Tom), V2 Run 1
04 / CRITICAL FINDINGS

TOP 3 INSIGHTS

Finding 1: The "Agents = Employees" Metaphor Backfired

Intended: Differentiate from tools (Klue/Crayon)
Reality: Created confusion and skepticism with tactical buyers

PMM IC (Sarah) - V1: 85% → V2: 71.5% (-13.5 pts)

V2 Confusion: "I get it's AI agents for competitive intelligence in Slack, but unclear what 'agents = employees' means in practice. Does it analyze/synthesize? Just alerts?"

Learning: Abstract metaphors fail with tactical buyers. They need concrete examples, not positioning.

VP Mktg (Tom) - V1: 73.5% → V2: 32.5% (-41 pts!)

V1 Success: "My team isn't going to log into another tool—they live in Slack. This could solve my adoption nightmare AND save budget."

V2 Failure: "After Crayon failure, 'agents automate everything' = empty promise. Needs '92% adoption in 30 days or full refund' or case study."

Learning: Burned buyers need proof > promises. V1's specificity ("Slack briefings") beats V2's abstraction ("agents do everything").

Finding 2: Value Chain Clarity = C-Level Gold

CRO (Elena) - V1: 32.5% → V2: 66.5% (+34 pts!)

V1 Confusion: "Can't tell if this is a tool FOR sales reps or FOR product marketing."

V2 Success: "Value chain is explicit: PMM uses tool → Reps get intel in Slack → Win rates up 22%. No ambiguity."

Learning: Exec buyers think in systems. Show the causal chain.

Finding 3: Pricing Transparency = Enterprise Unlock

VP Sales (Marcus) - Both: 63.5%

V1 Blocker: "$299/mo per seat or per team? 12 reps = $43K/yr vs $3.6K/yr = massive budget difference."

V2 Fix: "Wait, not per seat? So all 10 of my AEs get this for under $300? That's insane."

Learning: Enterprise buyers stop at pricing ambiguity. Clarify early.

05 / RECOMMENDATIONS

NEXT STEPS

1. Segmented Messaging Strategy

Serve different variants based on traffic source:

AUDIENCE
VARIANT
REASONING
C-level (CRO, VP Sales, RevOps)
V2
Value chain clarity critical
PMM ICs, mid-market
V1
Specificity > abstraction
Burned incumbents (Klue/Crayon)
V1
Need proof, not promises
PMs, self-serve buyers
V2
Autonomy framing works
Founders
Neither
Need entirely different messaging

Implementation: Use LinkedIn job title targeting, referral source, or firmographic data to route traffic.

2. Fix "Agents = Employees" Metaphor

Current (V2):
"LMTY agents are employees, not tools. They monitor, filter, synthesize, post automatically."

Better (Concrete Examples):
"LMTY agents work like a junior analyst:

  • Monitor 47+ sources (news, social, product pages, job posts, reviews)
  • Filter signal from noise (pricing changes, feature launches, exec moves)
  • Synthesize 3-bullet briefings ("Competitor X dropped Enterprise pricing 15%")
  • Post to Slack twice-weekly (or real-time for urgent moves)"

3. Add Proof for Burned Buyers

Add to V2:

  • Adoption guarantee: "92% team engagement in 30 days or full refund"
  • Case study callout: "Acme Corp had unused Crayon ($30K/yr). Switched to LMTY. Now 87% of marketing team actively uses insights."
  • Usage metrics: "Average customer: 156 Slack reactions/month on briefings = active engagement"

4. Test V3 Hybrid

Combine V2's value chain clarity with V1's concrete specificity. Replace "agents = employees" metaphor with bulleted workflow examples. Add social proof section with adoption metrics.

Expected lift: +10-15 pts across burned buyers, maintain C-level gains.