Deliverability Crisis at a Subscription Box
Crate & Root is a subscription box with 120,000 email contacts. The only recurring send is a weekly newsletter; 35% of contacts have not opened in 12 months. Open rate is 22%, click rate 0.3%, hard bounce 4%, and spam complaint rate 0.08%. Gmail delivery has degraded: 18% of Gmail sends land in spam in the last 30 days, up from 4%. The welcome popup offers 10% off but sends one email, then adds contacts to the main list. Apple Mail Privacy Protection affects 45% of opens, so open rate is inflated. The team wants to buy a 50,000-contact list and increase frequency to three sends per week to hit revenue targets. The lifecycle marketer has 60 days and no dedicated deliverability engineer. What should the team do?
1. Diagnose whether the problem is deliverability, engagement, or list quality. What data would you check first?
Read the worked answer
The symptoms mix list quality, engagement, and deliverability; they reinforce each other.
- List quality: 35% not opened in 12 months and 4% hard bounce indicate stale or invalid addresses. Hard bounce should generally be below 2%; 4% is high and hurts reputation.
- Engagement: 22% open is inflated by Apple Mail Privacy Protection on 45% of opens, so true human open rate is likely much lower. A 0.3% click rate is weak, but may be partly a relevance and frequency problem.
- Deliverability: Gmail spam placement rising from 4% to 18% is a serious reputation signal. A spam complaint rate of 0.08% is above the 0.05% danger zone many providers watch, though below 0.1%. Bounce and spam placement together suggest the list is being mailed too broadly.
- Acquisition: one welcome email then the main list means new subscribers get no onboarding and no expectation-setting. A 10% discount may attract discount-seekers who never engage.
Segment by domain (Gmail, Outlook, corporate), engagement recency (0-30, 31-90, 91-180, 181-365, 365+ days), source (popup, quiz, purchased), and bounce type. Check authentication (SPF, DKIM, DMARC), Gmail Postmaster Tools if available, spam complaint rate by campaign, and whether the last 30 days had a frequency spike. Do not buy the 50,000-contact list.
2. Propose a 60-day remediation plan with segmentation and send cadence.
Read the worked answer
Days 1-7 - stop the bleeding. Pause sends to 365+ day non-openers and hard-bounce addresses. Remove hard bounces and known spam traps. Reduce frequency to one high-quality send per week. Check authentication and unsubscribe headers. Do not buy a list.
Days 8-21 - segment and re-permission. Build segments: engaged 0-90 days, 91-180, 181-365, 365+, new subscribers, and Gmail versus non-Gmail. Send a re-permission message to 181-365 day non-openers asking whether they still want to hear from the brand, with a clear yes or no. Anyone who does not click yes enters a sunset flow and is suppressed after 30 days.
Days 22-45 - rebuild engagement. For engaged segments, send twice per week: one value or content email and one product or offer email. Personalize by purchase category and subscription status. For new subscribers, build a three-email welcome series: brand story, how to choose, first-order offer. Set expectations and ask for a preference or reply. Use clicks and site behavior, not opens, as engagement signals.
Days 46-60 - measure and scale. Reintroduce two to three sends per week only to the engaged segment if Gmail spam placement falls below 5% and complaints stay below 0.05%. Keep unengaged contacts suppressed. Monitor Gmail spam rate, bounce, complaint, click rate, revenue per recipient, and list growth. If Gmail placement does not improve, reduce volume further and investigate content and links.
3. The team wants to buy a 50,000-contact list and send three times per week. How should the marketer respond?
Read the worked answer
Recommend against both moves as proposed. A purchased list lacks consent and will likely contain spam traps, invalid addresses, and people who never asked for the brand. That would worsen the Gmail spam placement, increase complaints, and may create legal exposure under consent-based email laws. It would also dilute engagement data, because the team would be mailing people who will never open or click.
Three sends per week to a 120,000-contact list with 35% inactive contacts would accelerate complaints and reputation damage. Frequency is not a substitute for relevance.
Propose alternatives:
- Grow the list with consent: optimize the popup with a clearer value proposition, offer a quiz or welcome series, add referral incentives, and use onsite and social capture.
- Increase revenue per recipient first: segment by purchase category, build winback and replenishment flows, upsell subscription tiers, and test a two-email welcome series.
- Test frequency only on engaged contacts: compare two versus three sends per week on the engaged segment with a holdout. Measure revenue per recipient, complaint rate, and Gmail placement.
- Set a gate: do not increase frequency until Gmail spam placement is below 5% and complaints are below 0.05% for four weeks.
The marketer should present this as protecting revenue and sender reputation, not as resisting growth.