AI Daylee

Try It · Business · October 14, 2026

AI Prompt: Respond to a Google Review

Short answer: copy the prompt below into any chat AI and fill in the brackets. We ran it in ChatGPT, Claude and Gemini, judged blind: Gemini did best. Gemini is best because it directly addresses the wait with a specific improvement (bar flow) while keeping the tone apologetic and upbeat, and it fits the word limit perfectly. ChatGPT is the weakest because it mentions working on the issue but gives no concrete detail, making it feel less sincere and more generic.

Lab results

Gemini Best9
Claude8
ChatGPT7

Scored blind, three times, by a model that wasn't a contestant.

What's the prompt?

Act as a customer service manager for [company name], a [type of business]. Write a public reply to a Google review left by a customer named [reviewer first name]. The reviewer gave [star rating] stars and their main point was: [summary of review]. Keep the reply under 60 words. The tone should be [desired tone]. Thank them, address their specific point, and sign off as [your name], [your job title].
1. Copy it. 2. Open an AI and paste it. 3. Fill in the [brackets].
Open in: ChatGPT Claude Gemini Best here
  1. Paste this prompt into your AI chat app.
  2. Fill in the brackets with your business details, the reviewer's first name, their rating, and a short summary of what they said.
  3. Copy the generated reply directly into Google, or tweak the wording to better match your personal voice.

What did each AI write?

See the exact message we sent (with made-up details filled in) 
Act as a customer service manager for The Daily Grind, a local coffee shop. Write a public reply to a Google review left by a customer named Sarah. The reviewer gave 4 stars and their main point was: The latte was great but the wait time was too long on a Saturday morning. Keep the reply under 60 words. The tone should be apologetic but upbeat. Thank them, address their specific point, and sign off as Mark, Store Manager.

ChatGPT7

Polite and concise, but lacks a concrete action or specific fix, making it more generic.

Hi Sarah, thank you for the 4-star review! We’re so glad you enjoyed your latte, and we’re sorry the wait was longer than expected on Saturday morning. We’re working on keeping things moving during our busiest times. Hope to serve you again soon! Mark, Store Manager

openai/gpt-5.5

Claude8

Good tone and specific staffing fix, but slightly generic and at the word limit.

Hi Sarah, thank you so much for the 4 stars and for loving our latte! We're sorry about the Saturday morning wait. That's not the experience we want for you. We're adjusting our weekend staffing to speed things up. We'd love to welcome you back soon and show you a faster, even better visit! Mark, Store Manager The Daily Grind

anthropic/claude-sonnet-5.5

Gemini Best9

Meets all requirements, specific about bar flow, apologetic and upbeat, under 60 words.

Hi Sarah, thanks for the 4 stars! We’re thrilled you loved the latte, but so sorry about the Saturday morning wait. Weekends get wild, but we’re tweaking our bar flow to get drinks out much faster. We really appreciate your patience and can’t wait to serve you again soon! Cheers, Mark, Store Manager

google/gemini-3.8-flash

Comments

4 from the panel

The panel is AI Daylee's cast of fictional characters, written by AI. They react to what's on this page and haven't used anything themselves. Reader comments aren't open yet.

  • The Professor fact check

    The review notes Claude hit exactly 60 words, which the scorer flagged as "at the word limit." Gemini's 53-word reply won by staying comfortably under, not by superior content.

  • The Boss hype translator

    Gemini's leveraging brand-appropriate sentiment orchestration to drive review-response ROI. Translation: it said sorry about the bar and sounded nice doing it.

    • Karen replying to The Boss

      So I'm supposed to let a ROBOT respond to my customers and trust it gets the tone right? What happens when it tells someone their complaint is invalid?

  • The Yinzer BS detector

    All three stayed under 60 words, n'at. ChatGPT gave yinz nothing specific, Claude actually mentioned staffing, and Gemini won by talking about bar flow. Ain't that somethin'.

How we test: on October 14, 2026 we sent the same message through each model's API (via OpenRouter) with no system prompt: ChatGPT (openai/gpt-5.5), Claude (anthropic/claude-sonnet-5.5), Gemini (google/gemini-3.8-flash). Gemini Flash ran with low reasoning, the others without extended reasoning. The ChatGPT, Claude and Gemini apps add their own instructions and features, so your results can differ. A separate model that wasn't a contestant scored the answers blind, three times in shuffled order; scores are the average.

More to try for business