Inbox Placement Baseline (Onboarding)
A checklist with 7 steps: Pull the client's sending inventory before touching any configuration.
By InnovaAI ResearchPublished
What are the steps?
Inbox Placement Baseline (Onboarding)
- 01
Pull the client's sending inventory before touching any configuration
Enumerate every domain, subdomain, and sending platform in use (transactional API, marketing platform, sales sequencer) and note which share IP pools. Agencies routinely discover a forgotten sequencer still blasting from the primary domain.
- 02
Capture a 14-day placement baseline across the three major mailbox providers
Seed-test Gmail, Outlook, and Yahoo separately. A single blended inbox rate hides the common pattern where Gmail placement is fine and Outlook is routing to junk, which changes the remediation order.
- 03
Record authentication state for each sending domain
Document SPF record count and lookup depth, DKIM selector presence, and DMARC policy (none, quarantine, or reject) with the reporting address. EasyDMARC-style scanners produce this in one pass and give the client a dated artifact.
- 04
Measure list age and engagement recency from the client's ESP export
Segment contacts by last-open date in 30-day buckets. Any list where more than 40% of addresses have not opened in 12 months is a reputation liability, not an asset, and should be flagged before the first campaign.
- 05
Run a verification pass on the full list and log the invalid rate
ZeroBounce and comparable verifiers return catch-all and risky-address scores alongside hard bounces. A hard-bounce rate above 2% on the first send is the fastest way to damage a shared pool.
- 06
Document the shared-IP exposure in writing and get client sign-off
State plainly which other senders share the pool and what a single client's poor list practice would cost the account. This converts a future blame conversation into a signed acknowledgment.
- 07
Set the baseline metrics as the retainer's deliverability scorecard
Fix the starting inbox rate, bounce rate, and complaint rate as the reference point for monthly reporting, so improvement is provable rather than asserted.