AviaryAI: Keyword Research & Site Navigation

Overview

As part of AviaryAI's SEO remediation effort, keyword research and competitor analysis were conducted to inform both content strategy and site navigation restructuring. The findings reveal a significant organic opportunity: competitors have not meaningfully targeted use-case-specific keywords, leaving clear space for AviaryAI to establish early authority.

See the broader SEO remediation context in ^[1].

Current State

• Organic traffic: ~5 visits/month
• Indexed keywords: 2
• Root cause: Core pages contain fewer than 200 words of text — below Google's threshold for indexing. Pages must exceed ~300 words to avoid being treated as thin content.

Competitor Analysis

Competitors analyzed: Glia, Posh, Interface, L2P, Direct Link.

Key Findings

Competitive Opportunity

All competitors rank primarily for generic banking keywords (e.g., "banking AI") and inbound/brand terms. None have meaningfully targeted use-case-specific keywords — the precise area where AviaryAI's voice agent differentiation is strongest.

This creates a first-mover window: AviaryAI can rank for use-case queries (e.g., loan servicing voice agent, credit union AI assistant) before competitors build out that content.

Keyword Research Findings

Keyword volumes in this space tend to be modest (hundreds to low thousands of monthly searches) but carry high commercial intent — appropriate for a B2B product with a long sales cycle.

Keyword categories identified:

• Core product terms — e.g., "banking AI," "voice agent for credit unions"
• Use-case terms — specific workflows AviaryAI automates (highest opportunity, lowest competition)
• Industry vertical terms — credit unions, community banks, mortgage servicers

Long-tail, use-case-specific keywords are the priority for blog content, as they avoid cannibalizing future product/use-case landing pages while building domain authority.

Proposed Site Navigation

An initial navigation restructure was drafted based on keyword research. This is a starting point for iteration, not a final recommendation.

Proposed top-level structure:

Home
├── Products
├── Use Cases
├── Industries
├── Blog / Resources
└── [Safety & Security — existing, preserved]

Each of Products, Use Cases, and Industries has sufficient keyword volume to justify a dedicated section with indexable sub-pages.

Constraints

The following pages are off-limits for structural changes, per Justin Dwyer (AviaryAI):

• Homepage — messaging is intentionally focused on the voice agent wedge; structure to remain consistent
• Safety & Security page — content and structure to be preserved as-is

All other pages are fair game for restructuring, copy expansion, and SEO optimization.

Recommended Actions

Immediate (unblock indexing)

• [ ] Add 300+ words of text to all core pages — required before Google will index them
• [ ] Force a Google Search Console crawl of core pages after content is added
• [ ] Audit existing blog posts for length and keyword targeting

Short-term (build authority)

• [ ] Publish 1–2 keyword-targeted blog posts targeting long-tail use-case terms
• [ ] Ensure blog topics do not cannibalize planned use-case landing pages
• [ ] Finalize proposed navigation structure through a follow-up review with Justin

Medium-term (structural SEO)

• [ ] Build out Use Cases and Industries sections as keyword-justified landing pages
• [ ] Add security headers to the Webflow site (currently only STS policy is in place — a negative SEO signal)
• [ ] Develop a knowledge base product page that fits within the voice-agent-first narrative

Notes & Constraints

• The site is built on Webflow, which limits some technical SEO interventions (e.g., server-level config, SSH access). Standard PHP/server workflows do not apply; fixes must be made within Webflow's tooling.
• AviaryAI's brand narrative is deliberately focused on the voice agent as the primary wedge. The knowledge base and platform features should be surfaced without diluting that focus.
• The "Mortgage Forward" case study requires a non-data-driven format — they are a technology company and cannot provide quantitative outcome metrics. See ^[1] for case study format decisions.

Related

• ^[1]
• ^[2]
• ^[3]