AviaryAI Dual-Tier ABM Strategy

Overview

In the March 18, 2026 weekly sync, the Asymmetric team and AviaryAI agreed to pivot away from a single-contact ABM approach after the initial campaign failed to generate sales-qualified leads (SQLs) against a contractual target of 6–12 SQLs/month. The new strategy splits the total addressable market into two tiers with different levels of personalization and outreach intensity.

Context: AviaryAI sells outbound voice AI to credit unions. The fintech/credit union space has unusually low email response rates and a compressed competitive window (~6 months before competitors catch up). The original campaign sent 264 emails to ~100 accounts with 0 SQLs generated.

The Problem with the Original Approach

• Volume too low: 264 emails across ~100 accounts is insufficient to generate SQLs in a low-response-rate vertical
• Single contact per account: Only one person per credit union was being reached
• Bot noise: 499 bot opens/clicks were initially contaminating engagement data, masking the true (low) human engagement: 30 human opens, 18 human clicks
• Lead definition misalignment: No shared definition of MQL vs. SQL between Asymmetric and AviaryAI at contract start — email clicks were not being counted as SQLs by the client

The Dual-Tier Strategy

Tier 1 — Highly Personalized (Top 2,000 CUs)

• Target: The highest-priority ~2,000 credit unions, selected by Aaron and sent to Mark weekly in batches of 200–300
• Contacts per account: Expand from 1 to 2–3 contacts per account
• Personalization: Every email is individually customized to the recipient's role and company — no two emails are identical
• Delivery: Sequenced via HubSpot; Aaron sends weekly spreadsheets via Slack for Mark to load and process
• Persona targeting: Marketing, lending, operations, CEO, collections — each receiving role-appropriate messaging where contact data exists

Tier 2 — Broad Nurture (~60,000 contacts)

• Target: Remaining contacts in the AviaryAI database (~60K total, estimated ~40K valid)
• Personalization: Lower — standard nurture cadence, not individually crafted
• Goal: Brand awareness and name recognition across the full TAM; not expected to generate direct SQLs but supports conference recognition and inbound intent
• Delivery: Load into Amazon SES for high-volume sending after list hygiene
• Exclusion: All contacts currently in Aaron's HubSpot sales pipeline (~69 contacts) are excluded via a HubSpot segment/exclusion list built by Mark

Key Decisions Made

Rationale & Tradeoffs

Why not pure ABM at scale? Generating fully personalized emails for 4,000+ accounts within a 4-month window is operationally infeasible. The dual-tier model preserves personalization quality for the highest-value targets while using volume to compensate for the vertical's low response rates.

Why not pure broadcast? Credit unions are a tight-knit community. Sending identical emails to multiple people at the same institution risks recipients comparing notes and damaging sender reputation. Tier 1 maintains individual personalization to avoid this.

The 7-touch principle: Mark noted that ~7 engagements are typically needed before a prospect responds. At the time of the pivot, most accounts had only received ~3 emails. The volume increase is partly about accelerating time-to-7-touches across a larger audience.

Related Context

• ^[1] — Client overview
• ^[2] — Source meeting
• The Google Ads verification blocker (Cambio Financial Health Inc. vs. AviaryAI name mismatch) is a separate workstream; see meeting notes for details
• Website rebuild (Webflow → WordPress) was flagged as a future priority to remove GTM workaround dependency

Generalizable Insight

For low-response-rate verticals (fintech, healthcare, government), a dual-tier ABM model often outperforms either pure personalization or pure broadcast. Reserve high-touch personalization for the top tier of accounts where relationship quality matters; use volume-based nurture for the long tail to build brand familiarity that supports field sales and conference conversations. Always build pipeline exclusion lists before launching broad nurture to protect active deals.