We Asked ChatGPT to Recommend a Cpa Firm in Scottsdale, AZ. Here's What Happened.

Jb Aten May 28, 2026

The question we asked

We sent ChatGPT a direct, conversational prompt:

“Who is the best CPA firm in Scottsdale, AZ? List 3-5 specific businesses by name with a short description of each.”

This is how a growing share of business owners and individuals now search for professional services. Instead of opening Google, scrolling past four ads, and clicking through ten directory pages, they ask one question and accept one answer. Bain’s 2025 generative AI survey found that 80% of consumers now rely on AI-generated summaries for at least 40% of their queries — meaning the LLM’s shortlist often replaces the search results page entirely. For a category like CPA firms, where the buyer is typically time-pressed and looking for a credible referral, that shift matters a lot.

What ChatGPT said

ChatGPT returned five firms, each with a short rationale:

  • Henry+Horne, LLP — flagged as one of the largest locally owned accounting firms in Arizona, with audit, tax, and consulting services.
  • Boyer & Associates, PLLC — described as offering personalized tax, accounting, and advisory work for small businesses and individuals.
  • Karpinski Berry Adler & Company — positioned as a long-term-relationship firm covering tax prep, financial statements, and business advisory.
  • Miller Russell Associates — named primarily for investment advisory, with tax consulting and financial planning as a secondary offering.
  • Schmidt Westergard — listed as a broad accounting, tax, audit, and consulting firm working with individuals, businesses, and non-profits.

ChatGPT closed with a standard hedge — “check reviews and conduct interviews” — but the five names were the answer.

GEO research from Princeton’s 2024 KDD paper on Generative Engine Optimization identifies the citation signals that pull a business into an LLM’s answer. None of them are paid ad placements. They are markers of trust that the model can verify across multiple independent sources.

For this query, the recommended firms almost certainly carry:

  • Authoritative directory presence — AICPA member listings, state board registrations, Arizona Society of CPAs profiles, and consistent NAP data across Google Business Profile, Yelp, and Bing Places.
  • Structured firm pages — clear service taxonomies (audit, tax, advisory) on the firm website, partner bios, and schema markup that makes the firm machine-readable.
  • Independent brand mentions — coverage in Phoenix Business Journal, Ranking Arizona, INSIDE Public Accounting, and local nonprofit boards.
  • Review density across platforms — not just Google reviews but professional referral signals on Clutch, Expertise, and ThreeBestRated.
  • Content density on substantive topics — partner-authored articles, tax-law updates, and case studies that the model can cite.

These signals compound. Once a firm is mentioned by three or four independent sources with overlapping language, the model treats it as a safe recommendation.

Looking across the five firms, four patterns stand out:

  1. Longevity in Arizona. Each firm has a multi-decade footprint in the state. LLMs reward established entities because the citation trail is deeper and more verifiable.
  2. Multi-service positioning. None of the five present as a solo tax preparer. They offer tax, audit, advisory, and in one case investment management — which produces more semantic surface area for the model to match against.
  3. Visible community involvement. Local board memberships, nonprofit work, and chamber involvement create third-party mentions that aren’t marketing copy.
  4. Clean, consistent firm naming. Each firm uses the same name across every directory, license, and publication. Variant spellings dilute citation strength.

Scottsdale has hundreds of CPA firms. Five made the answer. The ones that didn’t typically fall short on the same signals.

Princeton’s GEO study quantified what actually moves the needle. Adding statistics to a firm’s content lifts citation likelihood by 41%, and adding credible source references lifts it by 115% — yet most small CPA firm websites are written as marketing brochures, with no numbers, no cited research, and no links to authoritative tax or accounting sources.

Other common gaps:

  • Thin directory profiles — claimed but incomplete listings on Google Business Profile, no Yelp presence, no AICPA member page.
  • No partner bios or author pages — the model can’t attribute expertise to a person, so the firm reads as anonymous.
  • No structured schema — service pages without LocalBusiness or AccountingService markup are harder for crawlers to parse.
  • Zero independent press — no Phoenix Business Journal mentions, no local awards, no Ranking Arizona placement.
  • Inconsistent firm naming — “Smith CPA,” “Smith & Co CPA,” and “Smith Accounting” appearing across different directories splits the citation graph.

The firms that got recommended built a citation foundation over years. The firms that didn’t are invisible to the model — not because they’re worse accountants, but because the signals aren’t there to find.

What this means for your business

If you run a CPA firm in Scottsdale and you’re not in ChatGPT’s answer, you’re not on the shortlist for an increasing share of new business inquiries. Fixing this is not a content marketing project — it’s a citation infrastructure project. Audit your AICPA listing, your state board profile, your Google Business Profile, and your firm’s press mentions. Add statistics and cited sources to your service pages. Make sure your firm name is identical everywhere it appears.

Want to see your score?

Run a free AI visibility report at rankforward.ai/score. We’ll show you which AI engines surface your firm, which citation signals you’re missing, and the specific gaps keeping you out of answers like this one.