Benchmark Templates

23 templates across 7 categories. Each template is fully editable β€” task types, criteria weights, and model shortlist are all pre-filled but overrideable.

πŸš€

Outreach & GTM

5 templates
πŸ“§
Cold Email Generation
cold emaillow risk

Generate personalised cold outreach emails for B2B SaaS sales, including opening hook, value prop, and clear CTA.

Suggested: gpt-4o, gemini-2-5-flash, claude-3-5-sonnet +2 more
🎯
Lead Qualification
lead qualificationlow risk

Score and qualify inbound leads based on firmographic and behavioural signals. Output a score and reasoning.

Suggested: claude-3-5-sonnet, gpt-4o-mini, gemini-2-5-flash +1 more
πŸ”­
Competitive Intelligence
prospect researchlow risk

Research a prospect company and surface competitive intelligence β€” recent news, pain points, buying signals β€” to inform outreach.

Suggested: claude-3-5-sonnet, gpt-4o, sonar-large +1 more
πŸ“‹
Proposal Generation
proposal generationlow risk

Draft a client-facing proposal or executive summary with a clear problem statement, proposed solution, value props, and pricing narrative.

Suggested: claude-3-5-sonnet, gpt-4o, llama-3-3-70b
πŸ”
Multi-Touch Outreach Sequence
cold emailcontent generationlow risk

Generate a 4-touch outreach sequence (cold email, follow-up, LinkedIn, phone script) for B2B enterprise sales.

Suggested: claude-3-5-sonnet, gpt-4o, gemini-2-5-pro +1 more
🏦

Financial Services

5 templates
πŸͺͺ
KYC Document Review
kyc reviewhigh risk

Review Know Your Customer submissions, flag missing or inconsistent fields, and classify completeness.

Suggested: claude-3-opus, gpt-4o, claude-3-5-sonnet +1 more
🚨
AML Transaction Screening
aml reviewcritical risk

Assess transaction patterns for AML red flags, classify risk level, and produce a structured SAR-ready summary.

Suggested: claude-3-opus, gpt-4o, llama-3-1-70b
πŸ“‰
Credit Risk Assessment
credit risk assessmenthigh risk

Evaluate a loan application against credit policy, identify risk factors, and produce a structured credit memo.

Suggested: claude-3-opus, claude-3-5-sonnet, gpt-4o +1 more
πŸ“Š
Earnings Call Analysis
summarizationreasoningmedium risk

Summarise an earnings call transcript, extract guidance, and flag risks or sentiment shifts.

Suggested: claude-3-5-sonnet, gpt-4o, gemini-2-5-pro +1 more
πŸ“œ
Regulatory Q&A
regulatory filinghigh risk

Answer regulatory questions using internal policy documents, flagging uncertainty and citing sources.

Suggested: claude-3-opus, gpt-4o, llama-3-1-70b +1 more
✍️

Content & Research

3 templates
✍️
Long-Form Content
content generationlow risk

Generate blog posts, whitepapers, or thought-leadership articles with clear argument structure and brand voice.

Suggested: claude-3-5-sonnet, gpt-4o, gemini-2-5-pro +1 more
πŸ”¬
Research Summarisation
summarizationlow risk

Condense research papers, reports, or long documents into structured executive summaries with key findings.

Suggested: claude-3-5-sonnet, gemini-2-5-pro, gpt-4o +1 more
πŸ—ΊοΈ
Market & Competitor Analysis
reasoningprospect researchlow risk

Analyse a market segment β€” size, trends, key players, and entry points β€” from provided context.

Suggested: gpt-4o, claude-3-5-sonnet, gemini-2-5-pro +1 more
πŸ’¬

Support & Success

2 templates
πŸ’¬
Customer Support Response
customer supportlow risk

Draft empathetic, accurate, and on-brand support responses for common and complex customer queries.

Suggested: claude-3-5-haiku, gemini-2-5-flash, gpt-4o-mini +1 more
πŸš€
Customer Onboarding Materials
content generationcustomer supportlow risk

Generate personalised onboarding guides, welcome sequences, and product walkthroughs tailored to customer segment.

Suggested: claude-3-5-sonnet, gpt-4o, gemini-2-5-flash +1 more
⌨️

Code & Engineering

4 templates
πŸ”
Code Review & Quality
code generationlow risk

Review a code snippet for bugs, security issues, code smells, and improvement suggestions.

Suggested: claude-3-5-sonnet, gpt-4o, o3-mini +2 more
⌨️
Code Generation
code generationlow risk

Generate production-ready code from a natural language specification, including tests and documentation.

Suggested: o3-mini, claude-3-5-sonnet, gpt-4o +2 more
πŸ§ͺ
Test Generation
code generationlow risk

Generate comprehensive unit and integration tests for a given code module, including edge cases.

Suggested: claude-3-5-sonnet, gpt-4o, o3-mini +1 more
πŸ“–
API & Code Documentation
code generationsummarizationlow risk

Generate clear, accurate docstrings, README sections, and API reference documentation from code.

Suggested: claude-3-5-sonnet, gpt-4o, gpt-4o-mini +1 more
βš–οΈ

Legal & Compliance

2 templates
βš–οΈ
Contract Clause Extraction
summarizationreasoninghigh risk

Extract and summarise key clauses from a contract β€” liability, IP, termination, SLAs β€” and flag non-standard terms.

Suggested: claude-3-opus, claude-3-5-sonnet, gpt-4o +1 more
πŸ”Ž
M&A Due Diligence Summary
summarizationreasoninghigh risk

Summarise a due diligence data room section β€” financials, legal, IP, or HR β€” and surface key risks.

Suggested: claude-3-opus, gpt-4o, llama-3-1-70b +1 more
πŸ₯

Healthcare & Life Sciences

2 templates
πŸ₯
Clinical Note Summarisation
summarizationhigh risk

Summarise clinical notes into structured patient summaries, extracting diagnoses, medications, and care plan.

Suggested: claude-3-opus, gpt-4o, claude-3-5-sonnet +1 more
πŸ’Š
Patient Communication Drafting
content generationcustomer supportmedium risk

Draft clear, empathetic patient communications β€” appointment reminders, care plan summaries, discharge instructions.

Suggested: claude-3-5-sonnet, gpt-4o, gemini-2-5-flash +1 more