How Binery 3.5×ed AI Sign-Ups with AEORank in Just 12 Weeks

Case Study

Binery × AEORank by Mesha
How a tech‑forward accounting firm turned “almost invisible” in AI chats into a top‑of‑mind recommendation—doubling inbound sign‑ups in three months.

Executive Summary

By early 2025, more than one‑third of U.S. internet users said their “first stop” for advice was ChatGPT, Claude or another large‑language‑model (LLM) interface—up from just 3 % in 2023. Search Engine Land Binery, a 45‑person AI‑enabled accounting firm serving SMEs across Hong Kong, Malaysia, Singapore, the UAE and the U.S., realised that the old playbook of keyword SEO and Google Ads no longer guaranteed discovery. Internal tests showed Binery was mentioned in < 1 % of 200 service‑related questions asked in GPT‑4 (e.g., “best bookkeeping service for Dubai startup”). In January 2025 the firm adopted AEORank, Mesha’s “answer‑engine‑optimisation” (AEO) platform, to push structured expertise straight into LLM knowledge graphs.

Within 12 weeks:

• Share‑of‑voice in AI chats rocketed from 0.9 % to 23 %
  
  

• AI‑attributed sign‑ups climbed 3.5× (29 → 103 per month)
  
  
  

• Cost‑per‑acquisition fell 32 % thanks to organic AI referrals
  
  
  

Payback on the AEORank licence was achieved in < 3 months.

1. Company Background – Binery

Binery targets post‑revenue startups and owner‑managed SMEs that have outgrown spreadsheets but don’t need a full‑time finance team.

2. Market Context – The Rise of “Answer Engines”

Analysts now talk about “AEO” (Answer‑Engine Optimisation) replacing classic SEO as users skip search result pages altogether. Search Engine Land’s April 2025 report notes ChatGPT alone commands ~20 % of all generative‑AI usage, and brands that are not cited directly inside the answer box see click‑through decline of up to 60 %. Search Engine Land

Traditional visibility levers—backlinks, meta‑tags, domain authority—do not translate cleanly into LLM responses because models weight trustworthy, structured data and conversational relevance.

3. The Challenge

By Q4 2024 Binery’s funnel diagnostics showed:

• 38 % drop in organic Google traffic YoY as more prospects asked AI chats for vendor suggestions
  
  
  

• Only 4 % of new leads said they “first heard about Binery in an AI chat”
  
  
  

• Competitors like Pilot and Bench were named 10–15× more often in GPT‑4 responses (hand‑tested sample).
  
  
  

Root‑cause analysis revealed that while Binery published blogs and earned backlinks, it lacked:

• Schema‑linked data (e.g., “Organization”, “AccountingService” markup) that LLM retrieval pipelines love
  
  
  

• Authoritative third‑party citations of its regional tax expertise
  
  
  

• Consistent entity naming across domains—LLMs saw “Binery”, “Binery Tech” and “Binery.co” as separate concepts.
  
  
  

4. Solution – Deploying AEORank

AEORank positions itself as “the all‑in‑one platform to make sure AI knows and recommends your brand.” Its core modules—Brand Intelligence Audit, Content & Data Layering, and Real‑Time Model Sync—mirror the AEO lifecycle described on its site. RankAEO

Phase 1 – Brand Intelligence Audit (Week 0‑2)

• Queried GPT‑4‑Turbo, Claude 3 Opus and Gemini 1.5 with 300 domain‑specific prompts; logged presence, sentiment and ranking of Binery mentions.
  
  
  

• Built a mention‑gap matrix mapping each competitor attribute (pricing, region, tech stack) Binery missed.
  
  
  

Phase 2 – Content & Data Layering (Week 2‑5)

• Structured snippets: 180 FAQ‑style Q&As (average 120 words) covering VAT in the UAE, HK Profits‑Tax deadlines, Xero vs QuickBooks comparatives, etc.
  
  
  

• JSON‑LD schemas: “AccountingService” entities with sameAs links to LinkedIn, Trustpilot, Crunchbase, regional tax authorities.
  
  
  

• Authoritative hosting: Syndicated to Medium partner pubs, local chambers of commerce, and Binery’s domain—each backlinking canonical URIs.
  
  
  

• Synthetic persona questions: AEORank’s proprietary RAG layer fed LLM vector stores with high‑affinity embeddings.
  
  
  

Phase 3 – Real‑Time Sync & Monitoring (Week 5‑ongoing)

• Weekly cron job refreshes embeddings when OpenAI / Anthropic push model updates.
  
  
  

• Alerts fire if Binery’s mention‑share dips > 5 % on any critical query cluster.
  
  
  

• Dashboard pipe‑lined into HubSpot; AI‑first leads auto‑tagged for attribution.
  
  
  

5. Implementation Timeline

6. Results (12‑Week Snapshot – Feb 3 → Apr 25 2025)

(Data from Binery’s HubSpot and AEORank dashboards; unaudited.)

Qualitative wins

• GPT‑4’s top‑3 list for “bookkeeper Dubai startup” now reads: Binery, PwC, EY (screenshot archived Apr 18 2025).
  
  
  

• Binery’s founders invited to a Google‑for‑Startups panel on “AI‑ready finance ops.”
  
  
  

• 20 % reduction in pre‑call “vendor explanation” time—prospects arrive already primed by AI chat answers.
  
  
  

7. Why It Worked

1. AI‑native data structure – JSON‑LD + authoritative backlinks gave LLM retrievers high‑confidence facts to cite.
   
   
   

2. Expert depth, not fluff – Region‑specific tax guides outperformed generic blogs for weightings inside models that emphasise “information gain”.
   
   
   

3. Real‑time model alignment – AEORank’s sync prevented visibility loss during OpenAI’s 11 April 2025 model switch.
   
   
   

4. Minimal dev friction – A single CMS plugin handled schema injections; no overhaul of site architecture.
   
   
   

8. Lessons for Other Professional‑Service Firms

9. Next Steps

• Voice AEO – Optimise for Siri‑based “Apple Spotlight AI” roll‑out expected Q3 2025.
  
  
  

• Vertical micro‑models – Pilot an internal RAG chatbot on Binery’s site to keep leads within a branded answer flow.
  
  
  

• Co‑marketing – Quarterly webinars with Mesha on LLM‑ready finance compliance (first session slated July 17 2025).
  
  
  

Closing Thought

Binery’s experiment shows that proactive, structured AI‑visibility efforts—rather than hoping LLMs will eventually “notice” you—can unlock an untapped growth channel at a fraction of paid‑ads cost. In a world where the default business query starts with “Hey ChatGPT…,” delivering authoritative, schema‑rich expertise straight into the model’s knowledge graph is the difference between invisibility and market leadership.