Skip to content

AI Cold Email Analyzer

Paste your subject and body. Nova scores it on the things that drive replies, tells you exactly what is weak and rewrites it so you can compare side by side.

  • 100% free
  • No signup
  • Powered by Nova AI

AI Cold Email Analyzer

Powered by Nova AI

Paste the body to start. The subject is optional but improves the review.

Your teardown appears here

An overall score, six dimension scores with specific feedback, what to fix and a rewrite you can copy.

How It Works

Three steps, no account needed

  1. 01

    Paste your subject and body

    Drop in the exact email you send, merge tags included. No sign-up, and nothing you paste is stored.

  2. 02

    Nova scores it like an SDR coach

    You get an overall score, six dimension scores (clarity, personalisation, value, CTA, length, spam risk) and blunt feedback that quotes the weak lines.

  3. 03

    Compare the rewrite and fix the issues

    Each issue comes with a specific fix. The rewrite keeps your intent, trims it under 90 words and gives it one soft ask. Copy it and test.

What makes a cold email get replies

Replies come from relevance and ease. Relevance means the first line is about the reader's situation, not your company. Ease means the ask is small: a yes/no question, not a 45-minute demo. Everything else, the social proof, the feature list, the signature banner, is there to support those two things or it should go.

The analyzer scores six dimensions because each one maps to a reason people ignore cold email. Unclear emails get skimmed. Generic emails get deleted. Emails with no proof get doubted. Big asks get postponed. Long emails get saved for later, which means never. Spammy emails never arrive.

  • Clarity: can the reader say what you do in one sentence?
  • Personalisation: is there a line only this person could receive?
  • Value proposition: is the outcome specific and believable?
  • Call to action: is it one question that takes seconds to answer?
  • Length: under 90 words, ask visible without scrolling?
  • Spam risk: any trigger words, capitals, links or images?

The lines that quietly kill reply rates

Some phrases are so common that readers stop at them. 'I hope this email finds you well' wastes the most valuable line in the email. 'My name is X and I work at Y' answers a question nobody asked. 'Businesses like yours' is a confession that you did not look at theirs. 'Just following up' admits the email has nothing new.

Claims without evidence are just as costly. '10x your pipeline' and 'industry-leading' cannot be checked, so the reader discounts them to zero. One concrete example (a number, a timeframe, a named outcome) is worth a paragraph of adjectives.

Weak:   We help businesses like yours 10x their pipeline.
Better: One agency onboarded 12 domains and 36 mailboxes in under ten minutes last week.

How to use the score

Treat the overall score as a direction, not a grade. An email scoring in the eighties is ready to test; anything in the fifties needs the top two issues fixed; below that, start from the rewrite. Fix issues in the order listed, since the first one usually explains most of the gap.

After editing, run the new version through the analyzer again and through the spam word checker. Then send it from a warmed-up domain with SPF, DKIM and DMARC in place. InboxOne sets those records automatically when it provisions a mailbox, so copy improvements actually reach the inbox.

  • Fix the subject line first; it decides whether the body is read.
  • Cut the opener to the reader's problem and the ask.
  • Replace every claim with one example.
  • Re-score, then A/B test the old and new version on a split list.

Frequently asked questions

Still stuck? Book a 30-minute deliverability call and we'll look at your setup together.

  • What does the cold email analyzer check?

    Six things: clarity, personalisation, value proposition, call to action, length and spam risk. Each gets a 0 to 100 score with feedback that quotes the weak phrase and suggests a replacement, plus an overall score, a verdict and a rewrite.

  • What is a good cold email score?

    Eighty or above means the email is ready to test. Between fifty-five and eighty, fix the top issues before sending. Below fifty-five, the structure is the problem; start from the rewrite rather than patching lines.

  • How long should a cold email be?

    Under 90 words for the first email and under 60 for follow-ups. The ask should be visible on a phone without scrolling. The analyzer scores length on its own, and the rewrite is always kept under 90 words.

  • Why is my email marked as high spam risk?

    Usually because of trigger words such as 'free', 'guarantee' or 'limited time', exclamation marks, capitals in the subject, or too many links and images. The feedback names the exact phrases. Run the full spam checker for a line-by-line pass.

  • Does the analyzer keep my email or use it for training?

    No. The text is sent to the model to produce the review and is not stored or used for training. Nothing is linked to an account, and there is no sign-up.

  • Can I analyse emails that contain merge tags?

    Yes. Paste them as they are, with {{firstName}} or similar tags. Nova treats them as personalisation placeholders and keeps them in the rewrite where they belong.

Get your first mailboxes live today.

Create an account, buy domains and mailboxes, and export them to your sending tool. Plans start at $39 a month for 10 mailboxes.