YourOutreachSUCKS
    Reference

    The Cold Outreach
    Scoring Rubric

    A complete, weighted rubric for evaluating cold outreach quality - based on the YourOutreachSUCKS offense taxonomy. Use it to score your own messages before sending, or to evaluate sequences systematically.

    Version 2.0April 2026youroutreachsucks.com/rubric
    // The grade scale

    What each grade means

    A
    8.0-10.0
    Exemplary
    Zero or one minor offense. Specific signal, precise ICP, real customer outcome, proportionate ask. Only 6% of scored messages reach this tier.
    • →Verifiable opening signal
    • →Named ICP in 3 dimensions
    • →One customer, one number
    • →Ask sized to value established
    B
    6.0-7.9
    Above average
    Strong fundamentals with one or two minor/major offenses. The research is real, the ICP is named, but something in the execution or close falls short.
    • →Good research, slight ask timing
    • →Named ICP, weak outcome claim
    • →Strong outcome, slight generic open
    C
    5.0-5.9
    Mediocre
    Some genuine effort visible but multiple tactical errors reduce effectiveness. Will occasionally work through volume, not quality.
    • →Good open, weak ICP and value prop
    • →Specific ICP, no customer proof
    • →Real signal, premature ask
    D
    4.0-4.9
    Poor
    Structural failures across multiple dimensions. Not broken enough to auto-fail but fundamentally unlikely to work at any meaningful rate.
    • →Generic open + vague ICP + buzzwords
    • →Template feel despite no merge fail
    • →Every element slightly wrong
    F
    below 4.0
    Failing
    Multiple critical failures. Often includes an auto-fail offense. The message is structurally broken and signals to the reader that no genuine effort occurred.
    • →Auto-fail from MERGE-FAIL or AI-WRITING
    • →3+ critical offenses simultaneously
    • →Template indistinguishable from spam
    // Score it yourself

    Interactive score calculator

    Check every offense present in your message. The score updates automatically. An Auto-fail caps the result at 2.0 regardless of other offenses.

    Check offenses present in your messageScore: 10.0 / A
    MERGE-FAIL ↗Auto-fail, caps score at 2.0
    -5
    AI-WRITING ↗Auto-fail, obvious AI-generated copy
    -5
    FAKE-RESEARCH ↗Generic compliment, no specifics
    -4
    VAGUE-ICP ↗"Companies like yours"
    -3.5
    PREMATURE-ASK ↗Calendar before value
    -3
    HUMBLE-BRAG-PIVOT ↗Compliment then immediate pitch
    -2.2
    PAIN-ASSUMPTION ↗Uninvited empathy, no evidence
    -2
    BUZZWORD-LOAD ↗Abstraction over specificity
    -1.8
    FOMO-BAIT ↗Manufactured urgency
    -1.8
    REPLY-ALL-ENERGY ↗Asking them to do discovery
    -1.5
    WALL-OF-TEXT ↗No structure, assumes commitment
    -1.8
    Estimated score
    10.0
    / 10 · A
    // Dimension-by-dimension

    How to evaluate each part of a message

    Opening - what earns the next sentence
    References a specific, verifiable, recent signal - something they posted, published, or did that proves the sender looked.
    +0 (no deduction)
    Generic compliment with no verifiable specifics. "Impressed by your background" with no background named.
    FAKE-RESEARCH -4.0
    Opens with flattery then immediately pivots to the pitch. The compliment is instrumentalized.
    HUMBLE-BRAG-PIVOT -2.2
    Assumes the recipient is experiencing a specific problem with no evidence they are.
    PAIN-ASSUMPTION -2.0
    Targeting - who this is actually for
    Names industry + company stage + specific role or problem. "Series B SaaS VP of Sales" excludes most companies. Correct.
    +0 (no deduction)
    "Companies like yours," "revenue teams," "businesses your size." No industry, no role, no specificity.
    VAGUE-ICP -3.5
    Execution - did QA happen
    All merge fields resolved. No [First Name], no [Company], no brackets visible in the final send.
    +0 (no deduction)
    Visible merge tag in the final message. Automatic fail regardless of all other content.
    MERGE-FAIL - auto-fail cap 2.0
    AI-powered, scalable, outcomes-driven, game-changing. Abstractions filling the space where specific claims should be.
    BUZZWORD-LOAD -1.8
    Close - what you're asking for
    Ask is proportionate to the value established. Low-friction alternative offered. "2-min video first if that's easier."
    +0 (no deduction)
    Calendar request before any value or relevance has been established.
    PREMATURE-ASK -3.0
    "Before your competitors do," "limited spots," "wrapping up outreach this Friday." Urgency manufactured without cause.
    FOMO-BAIT -1.8
    "Would love to learn more about your challenges." Asking the recipient to do discovery for free.
    REPLY-ALL-ENERGY -1.5
    Archive benchmarks
    5.4
    Average score of all filed cases. The baseline for "what actually gets sent."
    0.1
    Lowest score on record. 5 simultaneous offenses. Case #0038.
    9.5
    Highest score on record. Real signal, named ICP, customer outcome.
    4.46
    Average offense count per message. Failures compound.
    Quick pre-send tests
    →Could this opening line be sent to 100 people with the same job title? If yes, rewrite it.
    →Can you point to a specific thing they did, said, or published? If not, you have FAKE-RESEARCH.
    →Does your ICP description exclude most companies? If everyone qualifies, it's VAGUE-ICP.
    →Is there a customer name and a number in the value prop? If not, it's abstraction.
    →Have you QA'd every merge field? Open the preview. Read every word.
    →Is your ask proportionate to what you've established? If you haven't proven relevance, no calendar ask yet.

    This rubric is free to use for internal QA, training, or sequence review. If publishing results publicly, cite as:

    YourOutreachSUCKS Scoring Rubric v2.1 (2026). youroutreachsucks.com/rubric