outreach-scorer

Outreach Scorer: Pre-Send Quality Assessment — Complete Methodology

The Core Principle

Most cold emails fail before they're opened. The subject line is generic, the body is about the sender, the CTA asks for too much, and there's nothing personal enough to prove the email wasn't sent to 10,000 other people. The Outreach Scorer evaluates emails across 10 dimensions that predict reply rates, gives a numerical score, and provides specific rewrite guidance for every weak dimension.

The goal is not a "perfect" email. The goal is an email that clears the bar for a reply in the prospect's specific segment. An 8/10 email to a VP at a Fortune 500 looks very different from an 8/10 email to a founder at a 20-person startup.

---

The 10-Dimension Scoring Rubric

Each dimension is scored 0-10. The total score is the unweighted average of all 10 dimensions, giving a final score out of 10.

Dimension 1: Subject Line Specificity (Weight: HIGH)

The subject line determines open rate. It must be specific enough that the recipient believes the email was written for them — not pulled from a template.

Score	Criteria	Example
0-2	Generic, could be sent to anyone	"Quick question" / "Introduction" / "Synergy opportunity"
3-4	Category-specific but not personalized	"Improving your sales pipeline"
5-6	References their company or role	"Question about [Company]'s outbound motion"
7-8	References a specific, timely signal	"[Company]'s Series B — congrats + a thought"
9-10	References a specific observation with intrigue	"The gap between [Company]'s job postings and tech stack"

Improvement guidance for low scores:

• Add the company name or a person's name
• Reference a specific event, data point, or observation
• Create curiosity without clickbait — the email body must deliver on the subject line's promise
• Keep under 7 words. Mobile truncates at ~35 characters.

Dimension 2: Opening Line Relevance (Weight: HIGH)

The first sentence determines whether they keep reading. It must immediately prove you did your homework. Never open with yourself.

Score	Criteria	Example
0-2	Opens with self-introduction	"My name is [X] and I work at [Y]"
3-4	Opens with generic company praise	"I've been following [Company] and love what you're doing"
5-6	Opens with a relevant observation	"Noticed [Company] just expanded into [market]"
7-8	Opens with a specific, researched insight	"Saw your team posted 4 SDR roles this month — looks like outbound is a priority for Q2"
9-10	Opens with a specific insight connected to a problem	"With 4 SDR roles posted and no RevOps listing, your new reps are going to hit a data quality wall in about 60 days"

Improvement guidance for low scores:

• Delete any sentence that starts with "I" or "My"
• Reference something specific from LinkedIn, their website, job postings, or recent news
• Connect the observation to a challenge they likely face — show you understand the implication, not just the fact

Dimension 3: Personalization Depth (Weight: HIGH)

Beyond the opening line, how deeply is this email tailored to THIS specific person and company? Could you send the same email body to 50 other people with a find-and-replace on the company name?

Score	Criteria	Example
0-2	No personalization — pure template	Entire email is generic
3-4	Surface-level personalization	Company name inserted but body is generic
5-6	Role-aware personalization	Body references challenges specific to their title
7-8	Company-specific personalization	References their specific tech stack, team size, or business model
9-10	Situation-specific personalization	References a specific event + connects it to their unique context + shows you understand the implication

The personalization test: Cover the company name and person's name. Can you tell who this email was written for? If yes, score 7+. If no, score below 5.

Dimension 4: Value Proposition Clarity (Weight: HIGH)

Does the email clearly communicate what the prospect gains? Not what your product does — what changes for THEM.

Score	Criteria	Example
0-2	No value prop — just asks for a meeting	"Can we schedule 15 minutes?"
3-4	Feature-focused value prop	"We use AI to automate data entry"
5-6	Benefit-focused value prop	"We help sales teams save 5 hours/week on data entry"
7-8	Outcome-focused, segment-specific	"Series B sales teams that use us ramp new SDRs in 2 weeks instead of 6"
9-10	Outcome-focused with social proof and specificity	"We helped [similar company] cut SDR ramp time from 6 weeks to 13 days, which meant their 4 new hires were producing pipeline by week 3"

Improvement guidance for low scores:

• Replace "we do X" with "you get Y"
• Add a specific metric: time saved, revenue increased, cost reduced
• Add social proof: a named customer or a category ("Series B SaaS companies like yours")
• Make the value prop relevant to the specific signal or pain you identified in the opening

Dimension 5: Call-to-Action Quality (Weight: MEDIUM-HIGH)

The CTA determines whether they respond. Most CTAs ask for too much too soon.

Score	Criteria	Example
0-2	No CTA or multiple competing CTAs	Email just ends / "Let's schedule a call and I'll also send you our case study and pricing"
3-4	High-commitment CTA with no reason	"Are you free for 30 minutes this week?"
5-6	Medium-commitment CTA	"Worth a 15-minute conversation?"
7-8	Low-commitment, interest-based CTA	"Is this a challenge your team is facing? If so, happy to share how [similar company] handled it"
9-10	Binary, low-friction CTA tied to the value prop	"Would it be useful to see the specific workflow that cut [similar company]'s ramp time in half? Takes 10 minutes."

CTA principles:

• One CTA per email. Never two.
• First email CTA should be interest-based ("Is this relevant?"), not commitment-based ("Book a demo")
• Make it easy to say yes — reduce the ask
• Tie the CTA to the value prop — the CTA should be the logical next step after the value prop

Dimension 6: Length and Readability (Weight: MEDIUM)

Cold emails should be scannable in 8 seconds on a mobile phone. Most are too long.

Score	Criteria
0-2	Over 200 words, dense paragraphs, no white space
3-4	150-200 words, some paragraphs are too long
5-6	100-150 words, decent formatting
7-8	75-100 words, short paragraphs (2-3 sentences max), easy to scan
9-10	50-90 words, every word earns its place, visually clean on mobile

Target structure:

• Line 1: Personalized opening (1 sentence)
• Line 2-3: Bridge to value prop (1-2 sentences)
• Line 4: Value prop with proof (1 sentence)
• Line 5: CTA (1 sentence)
• Total: 4-6 sentences, 60-90 words

Improvement guidance for low scores:

• Cut any sentence that doesn't advance the email toward the CTA
• Remove adjectives ("innovative," "cutting-edge," "world-class")
• Split paragraphs longer than 2 sentences
• Read it on your phone — if you have to scroll, it's too long

Dimension 7: Tone and Authenticity (Weight: MEDIUM)

Does this read like a human wrote it for another human? Or does it read like a template, a ChatGPT output, or a marketing brochure?

Score	Criteria
0-2	Corporate jargon, buzzwords, overly formal ("I hope this email finds you well")
3-4	Professional but stilted — reads like a template
5-6	Natural tone but still feels like a sales email
7-8	Conversational, confident, reads like a peer wrote it
9-10	Reads like a note from a knowledgeable colleague — warm, direct, zero fluff

Red flag phrases that kill tone:

• "I hope this email finds you well"
• "I wanted to reach out because..."
• "We're an industry-leading..."
• "I'd love to pick your brain"
• "Touching base"
• "Circling back"
• "Just following up" (in a first email)
• "Revolutionary / game-changing / disruptive"

Dimension 8: Sender Credibility (Weight: MEDIUM)

Does the email establish why the recipient should listen to this specific sender? Not the company — the person.

Score	Criteria
0-2	No credibility signal — could be anyone sending this
3-4	Company name mentioned but no context
5-6	Company name + category ("We help sales teams with X")
7-8	Relevant social proof ("We work with [similar companies]")
9-10	Personal credibility + company credibility + specific relevance ("I've worked with 3 Series B sales teams on this exact problem, including [name]")

Improvement guidance for low scores:

• Add one specific customer name that the prospect would recognize as similar to them
• Reference a relevant piece of content you published
• Mention a mutual connection if one exists
• Your LinkedIn profile IS your credibility — make sure it backs up your email

Dimension 9: Follow-Up Intelligence (Weight: LOW-MEDIUM)

If this is email 2, 3, or 4 in a sequence, does it add new value or just repeat the first email with "just following up"?

Score	Criteria
0-2	"Just following up on my last email" with no new information
3-4	Slightly reworded version of the first email
5-6	New angle but still the same core pitch
7-8	New value — different case study, relevant article, or new signal
9-10	New signal-based hook ("Since I last wrote, [new development at their company]") + new value

Follow-up principle: Every email in a sequence should stand alone. If they only read email 3, it should still make sense and provide value.

Dimension 10: Deliverability Safety (Weight: CRITICAL — acts as a gate)

Technical factors that affect whether the email reaches the inbox vs. spam folder. A deliverability score below 5 is an automatic fail regardless of other scores.

Score	Criteria
0-2	Multiple spam triggers, HTML-heavy, 3+ links, tracking pixels
3-4	Some spam trigger words, 2 links, light formatting
5-6	Clean text, 1 link, no obvious triggers
7-8	Plain text, 0-1 links, no trigger words, clean sender reputation
9-10	Plain text, 0 links (or 1 naked URL), no triggers, warmed domain, verified DKIM/SPF

Auto-fail triggers (score the entire email 0 if present):

• Sending from a brand-new domain without warmup
• Including an unsubscribe link in a 1:1 cold email (signals bulk send)
• Using URL shorteners (bit.ly, etc.)
• HTML formatting in a cold email
• More than 2 links
• Image-heavy emails

---

Scoring Calculation

Raw Score

Average all 10 dimensions: (D1 + D2 + D3 + D4 + D5 + D6 + D7 + D8 + D9 + D10) / 10

Gate Check

If Dimension 10 (Deliverability) scores below 5, the overall score is capped at 3.0 regardless of other dimensions.

Final Interpretation

Score	Verdict	Action
8.0-10.0	Excellent. Send it.	Minor polish at most
6.5-7.9	Good. Sendable with tweaks.	Improve 1-2 lowest dimensions
5.0-6.4	Mediocre. Will underperform.	Rewrite opening and value prop at minimum
3.0-4.9	Poor. Don't send.	Major rewrite needed — start from scratch
0-2.9	Terrible. This is spam.	Complete rethink of approach

---

Segment-Specific Benchmarks

What "good" looks like varies by who you're emailing.

Enterprise (VP+ at 1000+ employee companies)

Dimension	Minimum Acceptable	Notes
Subject Line	7	Must be specific — they get 200+ emails/day
Personalization	8	They can smell templates instantly
Length	8	Under 75 words. They won't read more
Tone	8	Peer-to-peer, executive-level
CTA	7	Interest-based, not meeting-based
Target overall:	7.5+	Below this, don't send to enterprise

Mid-Market (Director/VP at 100-1000 employee companies)

Dimension	Minimum Acceptable	Notes
Subject Line	6	Company-specific is sufficient
Personalization	6	Role-aware personalization is acceptable
Length	7	75-100 words
Tone	7	Professional but human
CTA	6	Low-commitment meeting ask is fine
Target overall:	6.5+

SMB (Founder/Owner at <100 employee companies)

Dimension	Minimum Acceptable	Notes
Subject Line	5	Can be more direct — less noise in inbox
Personalization	5	Industry/role personalization is acceptable
Length	6	Can be slightly longer — they wear many hats and need more context
Tone	8	Must be authentic — founders despise corporate tone
CTA	7	Must be ultra-low-friction
Target overall:	6.0+

---

A/B Variant Scoring

When scoring two variants of the same email:

1. Score both emails independently across all 10 dimensions
2. Compare dimension-by-dimension
3. Recommend the higher-scoring variant
4. Identify which dimensions each variant wins on
5. Suggest a hybrid: take the best elements from each

Output format for A/B comparison:

## A/B Score Comparison

| Dimension | Variant A | Variant B | Winner |
|---|---|---|---|
| Subject Line | 7 | 5 | A |
| Opening Line | 6 | 8 | B |
| Personalization | 7 | 7 | Tie |
| Value Prop | 5 | 7 | B |
| CTA | 8 | 6 | A |
| Length | 7 | 6 | A |
| Tone | 6 | 8 | B |
| Credibility | 5 | 5 | Tie |
| Follow-Up Intel | N/A | N/A | — |
| Deliverability | 8 | 8 | Tie |

Overall: A = 6.6 | B = 6.7

Recommendation: Send Variant B. It has a stronger opening and value prop,
which are the highest-leverage dimensions for reply rate. However, swap in
Variant A's subject line and CTA, which are both stronger.

---

Rewrite Guidance by Dimension

When a dimension scores low, provide specific rewrite guidance — not just "make it better."

Subject Line (scored < 6)

Rewrite prompt: "Reference a specific signal, event, or data point about the prospect's company. Keep under 7 words. Create curiosity that the email body delivers on."

Opening Line (scored < 6)

Rewrite prompt: "Delete your current first sentence. Start with an observation about the prospect's company — something from their LinkedIn, job postings, news, or website. Connect it to a challenge they likely face."

Personalization (scored < 6)

Rewrite prompt: "You need more research. Check: their LinkedIn posts, their company's job postings, recent press, their tech stack (BuiltWith), their G2 reviews (if they're a vendor). Find one specific detail and weave it in."

Value Prop (scored < 6)

Rewrite prompt: "Replace features with outcomes. Instead of 'we do X,' write 'you get Y.' Add a number (time saved, revenue gained, cost reduced) and a comparable customer name."

CTA (scored < 6)

Rewrite prompt: "Reduce the ask. Replace 'Can we schedule 30 minutes?' with 'Is this a priority for your team right now?' or 'Would it be useful to see how [similar company] handles this?'"

Length (scored < 6)

Rewrite prompt: "Your email is too long. Target: 5 sentences, under 90 words. Cut any sentence that doesn't directly support the CTA. Remove adjectives, adverbs, and hedge phrases ('I think,' 'perhaps,' 'just')."

Tone (scored < 6)

Rewrite prompt: "Read your email out loud. If you wouldn't say it in conversation, rewrite it. Remove: 'I hope this finds you well,' 'I wanted to reach out,' 'I'd love to,' and any buzzwords."

Credibility (scored < 6)

Rewrite prompt: "Add one proof point: a customer name the prospect would recognize, a metric from a case study, or a relevant credential. If you have a mutual connection, mention them."

---

Anti-Patterns

1. Over-scoring personalization. An email that spends 3 sentences on personalization and 1 on value is out of balance. Personalization is the hook, not the meal. One well-placed detail is worth more than a paragraph of research regurgitation.

2. Optimizing for the wrong segment. An 8/10 enterprise email (terse, subtle, peer-tone) would score a 4/10 for SMB (too cryptic, needs more context). Always score relative to the target segment.

3. Ignoring deliverability. A beautifully written email that goes to spam is worthless. Deliverability is the gate check — pass it first, then optimize content.

4. Scoring without rewriting. A score is meaningless without actionable improvement guidance. Every dimension below 6 should come with a specific rewrite suggestion.

5. Template worship. Scoring a template and then blasting it to 500 people. The template is the starting structure. Personalization must be added per-prospect before scoring is meaningful.

6. A/B testing without statistical significance. Scoring two variants is useful for deciding which to send, but don't draw conclusions about "what works" from a sample of 2. A/B scoring is for pre-send optimization, not statistical testing.