Auditing Your Cold Email Program: A Checklist
When reply rates drift down over a few months, the instinct is usually to rewrite the templates first, because copy is the most visible and most editable part of the program. It is also, in practice, the least common actual cause of a decline. A proper audit works through domain health, list quality, copy, and cadence in that order, because problems upstream mask everything downstream and a copy fix cannot repair a deliverability problem. This checklist is the sequence to work through.
- Audit in order — domain and deliverability health first, then list quality, then copy, then cadence — because upstream problems make downstream metrics look broken even when they are not.
- A reply rate decline with stable or rising bounce and spam-complaint rates points to a copy or targeting problem; a decline alongside rising bounces or complaints points upstream to domain or list health.
- Most 'our copy stopped working' diagnoses turn out to be deliverability or list-quality problems in disguise, because a message that never reaches the inbox cannot be judged on its content.
- A full audit takes a few hours with the right tools and should be run quarterly, or immediately whenever core metrics move outside their normal range.
- The output of an audit should be a short, prioritized list of fixes ranked by expected impact, not a long document — the value is in the ranking, not the coverage.
Step 1: Domain and infrastructure health
Start here regardless of what symptom triggered the audit, because a domain or infrastructure problem silently caps every other metric downstream — no amount of copy improvement fixes a message that lands in spam.
Run through these checks in order; a failure at any point is worth fixing before moving to list quality, since it can make list-quality signals look worse than they actually are.
- SPF, DKIM and DMARC records: confirm all three are correctly configured and passing for the sending domain — a misconfiguration here is the single most common cause of a sudden deliverability collapse.
- Domain and IP reputation: check a reputation tool (several are free) for the sending domain and IP; a recent drop correlates strongly with recent list-quality or send-volume changes worth cross-referencing.
- Bounce rate: should sit under roughly 2–3%; a rising trend over the audit period, even if still under that threshold, is an early warning worth investigating before it crosses it.
- Spam complaint rate: should stay well under 0.1%; this is the metric mailbox providers weight most heavily, and even a small rise here can trigger inbox placement problems that look like a copy or targeting issue.
- Sending volume and pattern: check for sudden spikes or an inconsistent daily pattern, both of which read as suspicious to receiving mail servers relative to steady, warmed-up sending.
- Domain age and warm-up status: a domain or subdomain still within its first 4–8 weeks of sending should not be judged against benchmarks set for an established domain.
Step 2: List quality
Once infrastructure checks out, move to the list itself — a well-configured domain sending to a poorly matched or stale list will still underperform, and this stage often surfaces the real cause behind a reply-rate decline that looked like a copy problem.
The goal at this stage is separating a genuine targeting problem (wrong people) from a data-quality problem (right people, bad data) from a saturation problem (right people, already contacted too many times), since each has a different fix.
- Email verification freshness: confirm contacts were verified within the last 30–60 days, since email addresses decay — a list verified six months ago will show elevated bounces regardless of how it was built.
- ICP fit: sample 20–30 contacts and check them against your actual ideal customer profile criteria — a common finding is list scope drift, where a list built for one segment gets reused for outreach it was never targeted for.
- Duplicate and re-contact rate: check whether contacts are being re-sequenced too soon after a previous campaign, which depresses reply rate through fatigue rather than any problem with the current message.
- Source quality: for purchased or scraped lists, spot-check accuracy of firmographic data (title, company, industry) against LinkedIn or the company site — inaccurate targeting data undermines personalization even when copy is well-written.
- Suppression list hygiene: confirm unsubscribes, bounces, and prior 'not interested' replies are actually being suppressed from future sends — a leak here damages both deliverability and recipient trust.
Step 3: Copy and personalization
Only assess copy once domain health and list quality have been ruled out or fixed — reviewing copy first risks rewriting perfectly good messages in response to a problem they did not cause.
The review here is qualitative more than metric-driven, though a few numbers help identify where to focus attention within the copy itself.
- Subject line variety and specificity: check whether subject lines are specific to the recipient's likely situation or generic enough to read as templated — generic subject lines suppress opens and, on modern clients, correlate with lower reply rates too.
- Personalization depth versus swap-test failure: pull five random sent emails and ask whether the personalized line could be pasted into an email to a different company with one word changed — if yes, it is not real personalization.
- CTA sizing: confirm the ask matches the trust level earned by that point in the sequence — a first-touch meeting request is a common, easily fixed cause of low reply rates.
- Length and readability: cold emails over roughly 150 words start losing reply rate on mobile-first readers; check actual average length against this rough ceiling.
- Claim accuracy: verify any statistic, client reference, or capability claim in current templates is still accurate — stale claims (an old client logo, an outdated statistic) erode trust with informed readers.
- Reply rate by variant: if A/B testing is running, confirm a variant is actually statistically distinguishable before drawing conclusions from it — small-sample 'winners' are a common source of bad copy decisions.
Step 4: Cadence and sequencing
The final layer is timing and structure — how many touches, how spaced, and in what order — which can undermine even strong copy sent to a clean list.
Cadence problems are often the least visible cause of underperformance because the individual emails all look fine in isolation; the problem only shows up in aggregate sequence-level data.
- Touch count and spacing: confirm the sequence length (commonly four to six touches) and spacing (typically two to four days apart) match what has worked historically for this list type — sequences that are too short abandon contacts before a realistic reply window; too long risks fatigue and complaints.
- Day-of-week and time-of-day pattern: check whether sends cluster at times with historically weaker open and reply performance for the target vertical.
- Angle variation across touches: confirm later touches in a sequence offer a genuinely different angle rather than restating the first email with 'just following up,' which is a common, easily fixed cause of declining reply rates deeper into a sequence.
- Multi-threading presence: for high-ACV accounts, check whether the sequence reaches more than one stakeholder, since single-contact sequences underperform on deal survival even when reply rate looks fine.
- Stop conditions: confirm the sequence correctly stops on a reply, bounce, or unsubscribe rather than continuing to send — a broken stop condition is both a compliance risk and a fast way to generate spam complaints.
Reading the results together
The pattern across metrics tells you more than any single number. A reply rate decline with stable bounce and complaint rates points downstream, toward copy or targeting. A reply rate decline alongside rising bounces or complaints points upstream, toward domain or list health — and fixing copy first in that scenario wastes effort on a symptom rather than the cause.
A useful gut check: compare the current period's numbers against the same program's own numbers from three to six months ago, not against generic industry benchmarks. Internal trend comparison isolates what changed in your specific setup, while external benchmarks mix in variables (vertical, offer, list source) that have nothing to do with what actually shifted in your program.
A program showing reply rate down from 6% to 3%, bounce rate stable at 1.5%, and spam complaints stable near zero is almost certainly a copy or targeting issue — check swap-test failures and CTA sizing first. The same reply-rate drop alongside bounce rate rising from 1.5% to 4% points straight at list quality — check verification freshness before touching a single template.
Turning the audit into action
Resist the temptation to fix everything the audit surfaces at once — changing domain settings, list sourcing, copy, and cadence in the same week makes it impossible to know which change produced which result. Prioritize fixes by expected impact and implement them in the same order the audit ran: infrastructure first, then list, then copy, then cadence, measuring each change's effect before layering on the next.
Schedule the next full audit for roughly a quarter out, and set up lightweight ongoing monitoring — bounce rate, complaint rate, and reply rate on a weekly dashboard — so the next real problem gets caught within weeks rather than the months it typically takes for a slow decline to become obvious enough to prompt a full audit.
FAQ
What should I audit first when reply rates drop?
Domain and deliverability health — SPF/DKIM/DMARC configuration, domain reputation, bounce rate, and spam complaint rate. Problems here silently cap every downstream metric, so ruling them out first prevents wasted effort rewriting copy that was never the actual cause.
How do I tell if a reply rate decline is a copy problem or a deliverability problem?
Check bounce rate and spam complaint rate alongside the reply rate. If those are stable, the issue is likely downstream in copy or targeting. If they are rising too, the issue is upstream in domain or list health, and copy fixes will not resolve it.
How often should a cold email program be audited?
Roughly quarterly as a baseline, plus immediately whenever a core metric — reply rate, bounce rate, or complaint rate — moves outside its normal range unexpectedly.
What is the most common cause of underperformance that audits reveal?
List quality issues — stale verification, ICP drift, or re-contacting the same saturated list too often — are found more frequently than genuine copy problems, even though copy is usually the first thing teams suspect and rewrite.
Should I fix everything an audit finds at once?
No. Fix issues in the order they were audited — infrastructure, then list, then copy, then cadence — and measure the effect of each change before making the next one. Changing everything simultaneously makes it impossible to know what actually worked.
What tools do I need to run this audit?
A domain reputation checker, your email verification tool's activity log, your sending platform's bounce and complaint reports, and access to a sample of sent emails and sequence configuration. No specialized audit software is required for most of this checklist.
Want to apply this to your outreach?
We will map it to your segment and product — before any work starts.
Talk to us