Your Outbound Isn't Working. Here's How to Diagnose It in 30 Minutes.

Every week someone DMs me on LinkedIn: "Cold email doesn't work for us." Ninety percent of the time, I can find the problem in under 30 minutes. And ninety percent of the time, it's not the emails.

Here's the diagnostic I run.

Step 1: Check Your Deliverability (5 Minutes)

If your emails aren't reaching inboxes, nothing else matters. Check bounce rate (should be under 3%), domain health, and SPF/DKIM/DMARC records. New domains need 14-21 days of gradual warmup before sending volume.

Before you touch anything else. If your emails aren't reaching inboxes, nothing else matters.

Check bounce rate first: above 3% means bad data, above 5% means your domain reputation is taking damage with every send. Use a deliverability testing tool (mail-tester.com, GlockApps). Send a test email and see where it lands. If you're hitting spam, stop everything else.

Fix infrastructure first. Check domain age, warmup status, SPF/DKIM/DMARC records. If you haven't done domain warming yet, you're sending emails from a domain that Gmail doesn't know. It lands in spam. It doesn't matter if your email says "I have a special offer for you." It's not reaching anyone.

The 5 Things That Kill Outbound Campaigns post has the full warmup section if you need to set that up.

Step 2: Check Your Targeting (10 Minutes)

Pull your last 200 prospects. Ask: Would 70%+ genuinely benefit right now, not someday? Look at who replied and find the patterns. The answer is almost always a situational trigger you weren't filtering for.

Pull up your last 200 prospects. Ask yourself: Would every person on this list genuinely benefit from what I sell? Not "could they theoretically use it someday." Would they benefit right now?

If you can't answer yes for at least 70% of the list, your targeting is the problem.

Look at reply patterns. Who actually responded? What do those companies have in common that the non-responders don't? The answer is almost always a situational trigger you weren't filtering for. Maybe it's company size, maybe it's industry, maybe it's recent funding or a job opening. But there's a pattern in the yeses.

Go find that pattern. Then filter your list to only people matching it. This is why defining your ICP for outbound matters. Not because it's a nice-to-have exercise. Because narrower targeting = higher reply rate = real data you can act on.

Step 3: Check Your Messaging (10 Minutes)

Only check messaging after deliverability and targeting pass. Read your email out loud. Does it sound human? Or corporate? The #1 killer: leading with your product instead of their problem.

Only check this after deliverability and targeting pass. Read your first email out loud. Does it sound like a person talking to another person? Or does it sound like a marketing department wrote it by committee?

The #1 messaging killer: leading with your product instead of their problem. First line should reference something specific about the prospect's situation. If your email could be sent to 10,000 people without changing a word, it's not personalized enough.

Real personalization isn't mail merge. It's not "Hi {{first_name}}, I noticed {{company_name}} does {{thing}}." That's generic with variables. Real personalization is: "I saw your company just raised Series B funding, you're hiring like crazy, and your customer base is mostly Fortune 500 companies. That suggests scaling ops is a pain point."

If you want to understand the difference, read AI Personalization vs Mail Merge. The tools matter here too, but the thinking comes first.

Step 4: Check Your Volume and Patience (5 Minutes)

Under 500 emails = not enough data. Under 3 weeks = haven't seen follow-ups yet. Most meetings come from email 2-3, not 1. You need volume before you can diagnose anything else.

How many emails have you actually sent? If it's under 500, you don't have enough data to diagnose anything else. How long has the campaign been running? If it's under 3 weeks, you haven't even seen your follow-up sequences play out.

Most meetings come from email 2 or 3, not email 1. Quick math: 500 emails × 8% reply rate = 40 replies. 40 replies × 50% positive = 20 conversations. 20 conversations × 30% meeting rate = 6 meetings.

If you sent 100 emails and got 1 meeting, that's normal. That's actually fine. You just need more at-bats to see the real pattern. If you've run 500+ emails, seen 3+ weeks of data including follow-ups, and your reply rate is still under 2%, then yes, something is wrong. But you need volume first.

Track the right metrics. Want to know what those are? Outbound Sales Metrics That Actually Matter breaks down what to measure and why.

The Decision Tree

The diagnostic order matters. Check deliverability → targeting → messaging → volume. Stop at the first problem and fix it before moving to the next. Most founders skip straight to copy and miss the real issue.

Put this in a doc. Refer to it monthly. The diagnostic order is the whole point.

Bounce rate > 5%? Fix your data. Verify emails before sending. Use Hunter, Clearbit, RocketReach. Bad data is expensive.

Landing in spam? Fix infrastructure. Warm your domains. Check authentication. Wait. Don't send yet.

Reply rate < 2%? Targeting is off. Tighten your ICP. Add urgency signals. Look for situational triggers in your replies.

Reply rate 2–5% but all negative? Messaging is off. Lead with their problem, not your product. Make it about them first.

Reply rate 5%+ but few meetings? Your CTA or follow-up is weak. Or your qualifying criteria are too tight. You're talking to the right people but asking for the wrong next step.

Everything looks fine but no pipeline? Check volume. You probably need more at-bats. Or the time horizon is too short.

The Mistake Everyone Makes

Founders rewrite email copy first. That's backwards. The issue is almost always upstream: deliverability, targeting, or volume. Copy is the last variable to touch, not the first.

The first instinct is always to rewrite the email. I get it. It's the most visible, most controllable variable. You see a 2% reply rate and think: "I need a better subject line."

But changing subject lines when your domain is landing in spam is like repainting a car that has no engine. The diagnostic order matters: deliverability → targeting → messaging → volume. Always in that order.

Spend 3 weeks A/B testing subject lines and you've wasted 3 weeks. The real issue was sitting upstream the whole time. You were emailing the wrong people from a domain that Gmail doesn't trust.

Most outbound problems are upstream problems. Solve them in order.

How AI Changes This (But Doesn't)

AI tools improve messaging quality and speed, but they can't fix upstream problems. Bad data stays bad. Wrong targeting stays wrong. AI is a force multiplier, not a substitute for thinking.

If you're using AI to research prospects and personalize at scale, the diagnostic is the same. The deliverability check is the same. The targeting check is the same. The volume check is the same.

What changes: your messaging quality and speed. AI tools can personalize faster. They can research better. They can iterate faster. But they can't fix upstream problems. If your data is bad, AI won't fix it. If your targeting is wrong, AI won't fix it.

AI is a force multiplier, not a replacement for thinking. How AI Outbound Actually Works digs into where the real value is.

Run This Monthly

Outbound systems fail slowly, not suddenly. Reply rates drift 5% → 4% → 2.5% before you notice. Block 30 minutes monthly to run this diagnostic. It'll save weeks of wasted optimization.

Most outbound systems don't fail suddenly. They fail slowly. You ship 50 emails a day. Reply rate drifts from 5% to 4% to 2.5%. By the time you notice, you've wasted weeks on the wrong problem.

Block 30 minutes on your calendar once a month. Run this diagnostic. It takes 30 minutes. It'll save you from the slow death spiral of optimizing the wrong thing.

Related: 5 Things That Kill Outbound Campaigns in the First 30 Days • Outbound Sales Metrics That Actually Matter • How to Define Your ICP So Outbound Actually Works