You've built a cold email campaign. You've got solid copy. But your reply rate is sitting at 1.2% when it should be 4-6%, and you're wondering why.
Most of the time, it's not the email. It's the data.
Bad email data kills campaigns silently. You're sending to the wrong people, wrong titles, wrong company sizes, or wrong emails entirely. Your open rates look fine (because some emails are getting through), but nothing converts because you're talking to the wrong person in the wrong context.
Here's how to diagnose and fix it.
The Three Data Problems That Kill Cold Email
Before you can fix data problems, you need to know what you're looking for. There are really three categories of bad data, and each one requires a different fix.
Problem 1: Wrong Person
You're emailing someone with the title "Manager" at a company with 200 employees. In a company that size, a Manager might be an individual contributor managing a team of 2 people. They don't have budget authority. They don't make the call. They can't say yes.
This is the most common data problem. Your lead list has real people at real companies, but you're targeting the wrong seniority level, the wrong department, or both.
How to diagnose it: Look at your bounce rate and your reply rate separately. If bounce rate is low (under 2%) but reply rate is also low (under 3%), you've got the right email addresses going to the wrong people.
Problem 2: Wrong Email Address
The email address exists, but it's not the right email for that person anymore. They moved to a different company 6 months ago. They got promoted and now use a different address format. They're on parental leave and the email is forwarded to someone else.
This shows up as a higher bounce rate - typically 3-5% is normal, but if you're seeing 8-12%, something's wrong with your email validation.
Problem 3: Wrong Company
You're sending to companies that don't fit your ideal customer profile. Maybe they're too small, too large, wrong industry, wrong geography, or they've already bought a solution from a competitor.
This one's harder to spot in the data because the emails might deliver fine, but you'll see extremely low reply rates (under 1%) across the board, and when you do get replies, they're mostly "not a fit" or "we already use someone for this."
How to Audit Your Data Quality Right Now
You don't need fancy tools for this. You need 30 minutes and a spreadsheet.
Step 1: Pull a random sample of 50 records from your list. Not the first 50 (those might be better validated). Pick every 20th record or use a random number generator.
Step 2: For each person, verify three things:
- Is this person's LinkedIn profile still active at this company? (Check LinkedIn directly - don't rely on what your tool says)
- Does their title and seniority level make sense for your offering? Would they actually care?
- Is this company within your ICP? (Size, industry, location, etc.)
Step 3: Count how many pass all three checks. If fewer than 35 out of 50 pass, you have a serious data quality problem. If it's 40+, your data is decent.
This gives you a data quality score. If you're at 70% data quality, you can expect about 70% of the upside in your results.
How to Fix Each Type of Data Problem
Fixing "Wrong Person"
This requires three changes to how you source and filter leads.
First, tighten your title filters. Don't use "Manager." Use "VP of Sales," "Director of Marketing," "Head of Operations," or "CFO." These are people who actually make decisions.
If your tool allows it, use include filters, not exclude filters. Instead of "exclude anyone with 'intern' in their title," say "include only people with 'VP,' 'Director,' 'Manager,' or 'Head' titles" - but not generic Manager. Be specific about what Manager means in your context. For a small company, a Manager might be a decision-maker. For a 500-person company, they might not be.
Second, verify company size matters for you. If you're selling to mid-market (50-300 employees), companies with 5 people and companies with 5,000 people are both wrong, but for different reasons. Be explicit about your target company size range and stick to it.
Third, filter by department intent signals. If you're selling HR software, don't just email the VP of HR. Email the VP of HR at companies where HR is clearly a priority - that might mean companies that recently posted 5+ open HR roles, or companies in high-growth industries, or companies with public hiring announcements.
Fixing "Wrong Email Address"
Bad email addresses typically come from one of two sources: outdated data or poor validation.
For outdated data: Before you send a campaign, run your list through email validation. Services like ZeroBounce or NeverBounce cost $50-200 for a 10,000-record validation. They'll flag bad addresses before you send to them, which protects your sender reputation and saves wasted sends.
If your bounce rate is already above 5% on an active campaign, pause it. Pull out all the bounced emails and remove them from future sends.
For poor validation: Look at where you're sourcing emails from. Tools like Apollo and lemlist have different data quality. If one tool is giving you higher bounce rates than another, switch sources for email addresses only, even if you're using the other tool for lead research.
Fixing "Wrong Company"
This is the hardest to fix because it requires rethinking who you're targeting.
Start by looking at your best customers - the ones who actually signed. Pull 3-5 of them and write down exactly who they are. Not "mid-market SaaS companies." Specific: "Series B SaaS companies in marketing automation, $2M-5M ARR, based in US, founded in last 5 years."
Now rebuild your target list to match that profile as closely as possible. This might mean getting more specific with geography, funding stage, or growth signals.
Use LinkedIn Sales Navigator filters to layer in intent signals - companies that recently posted jobs in your target role, or recently announced funding, or updated their company description. These companies are active and making changes, which means they're more likely to be receptive to what you're selling.
The Numbers That Matter
Once you've fixed your data, here's what healthy metrics look like:
- Bounce rate: 1-3% is good. Above 5% means bad emails or bad data validation.
- Open rate: 25-35% is typical. Below 15% usually means subject line or deliverability issues, not data issues.
- Reply rate: 3-6% is solid. Below 2% usually means wrong person, wrong message, or wrong company.
- Data quality score (from your manual audit): 70%+ is acceptable. 85%+ is good. Below 60% means you need to completely rebuild your list.
If you fix your data and still see low reply rates, the problem is your message, not your audience. That's actually better - copy is easier to fix than data.
When Data Becomes a Full-Time Job
Once you understand what good data looks like, you realize maintaining it at scale is a different beast entirely. Sourcing clean lists, validating addresses, removing bounces, updating old records, segmenting by intent signals - it's doable solo for 500-1000 leads, but at 5,000+ it becomes a logistics problem. Some teams bring in help to handle the data layer while they focus on copy and follow-up strategy, which lets the whole campaign run tighter.
Related Guides
- How to Find Email Addresses for B2B Cold Email (Without Losing Your Mind)
- How to Build a Cold Email List From Scratch (Without Losing Your Mind)
- How to Track Cold Email Campaigns (So You Actually Know What's Working)
- How to Write Cold Emails That Actually Get Replies
- How to Book 20 Meetings a Month with Cold Email (Without Losing Your Mind)