TL;DR. Run the POC on a dedicated sister domain (not your primary corporate domain). Provision 2 to 4 managed mailboxes and warm them for 21 days before any cold sends. Document four success thresholds in week 1: bounce rate under 3 percent, open rate at or above 25 percent, reply rate at or above 2 percent on cold signals, and one qualified opportunity per 250 enrollments. Per the Unify Managed Deliverability product page, the platform handles domain creation, automated 21-day warm-up, and pre-send bounce validation. Per the Justworks case study, this approach prevented over 10 percent of bounces in outbound enrollments at production scale.
Methodology and limitations
How the 21-day warming window is measured. Per the Unify Managed Deliverability product page, mailbox warming runs automatically over a 21-day ramp period (described as "up to 3 weeks" on the same page). Warming gradually increases send volume per mailbox while monitoring open and reply behavior to build sender reputation with Google and Microsoft. The clock starts when the mailbox is first provisioned and authenticated, not when cold sends begin. Cold sends should not start until day 21 even if early signals look healthy.
Customer outcomes are named, not aggregated. Every quantitative POC outcome cited in this article is attributed to a specific named customer case study. There is no aggregated "Unify POC benchmark" dataset. Dial benchmarks down when your ACV is under $25K, your ICP is unproven, your CRM data is dirty, or you operate in regulated regions (EU, financial services, healthcare) where consent workflows extend the timeline. Dial benchmarks up when ACVs exceed $100K, ICP is well-defined, and CRM is clean.
How does email management work during an outbound platform POC?
Buy a dedicated sending domain, provision 2 to 4 managed mailboxes on it, and warm them for 21 days before any cold sends. Never pilot on your primary corporate domain. Per the Unify Managed Deliverability product page, the platform handles domain registration, automated DNS configuration, mailbox setup, 21-day warming, and pre-send bounce validation as a single managed flow. Per the Innovate Energy Group case study, this end-to-end approach replaced prior warming services that "used artificial engagement pools and damaged domain reputation, causing messages to land in spam."
The POC starts on day one with infrastructure, not sends. CRM authentication, DNS setup (SPF, DKIM, DMARC), and domain warming run in parallel during the first three weeks. The first cold sends go out around day 21 to 22, leaving 8 to 10 days of measurable send volume in a typical 30-day POC. Plan the calendar with the warming clock as the constraint, not the kickoff date.
Should we use a subdomain or a separate domain for the POC?
Use a separate, parked domain that closely resembles your primary domain. Examples: company-mail.com, get-company.com, or company-outbound.com if your primary is company.com. A subdomain (such as outbound.company.com) shares root-domain reputation, which means a deliverability incident on the subdomain can damage your corporate email reputation. A sister domain isolates pilot risk entirely.
The exception is enterprise tiers with mature SPF, DKIM, and DMARC alignment, high baseline sender reputation, and IT review of the subdomain setup. Even there, the cost of registering a sister domain ($10 to $20 for the registration) is so much smaller than the cost of a corporate-email reputation hit that the default should be a sister domain. Per the Affiniti case study, Unify auto-creates and warms a domain and inbox at setup, so the operational cost of running a sister domain is near zero.
Subdomain strategy: parking vs live
Park the domain at registration. Configure DNS records (SPF, DKIM, DMARC) and the inbound forwarding rule (replies to the sending domain forward to a monitored corporate inbox) within the first 48 hours. Do not point a website at the sending domain. The cleanest configuration is a domain that serves only as an email-sending entity. If the domain serves marketing pages or trial signups, the reputation signal becomes harder to interpret.
How many mailboxes do we need for the POC?
Two to four for a 30-day POC at typical SMB or mid-market volumes (under 5,000 sends in the pilot window). Each mailbox can send roughly 30 to 50 cold emails per day after warming completes, so the math works out as: 4 mailboxes x 40 sends/day x 10 days of active sending = 1,600 sends in the live window. That sample is large enough to measure bounce rate, open rate, and reply rate against thresholds.
Per the Unify pricing page, 5 managed Gmail mailboxes are included on the Growth monthly tier (a typical POC tier), 8 on Growth annual, 20 on Pro, and 40 on Enterprise. Additional mailboxes are $25 per mailbox per month. Avoid over-provisioning: idle mailboxes during the POC waste warming cycles, and unused capacity is a vanity provisioning metric.
How do you isolate sandbox vs production data without breaking attribution?
Two safe configurations. Option A: connect the POC to a Salesforce or HubSpot sandbox instance, isolate the audience to a defined account list, and treat all attribution as pilot-only. Option B: connect the POC to production CRM but restrict enrolment to a flagged audience segment, with exclusion rules preventing the POC from touching active sales-owned accounts.
Option B is the more common choice because attribution flows to real opportunities and pilot-to-production handoff requires no data migration. The trade-off is that production-CRM POCs require disciplined exclusion rules and routing logic. Per the Unify Salesforce integration page, the platform supports selective record syncing with customizable exclusions, custom field utilization, and lead routing — the mechanics needed for safe production-CRM POCs.
Required isolation mechanics
- Exclusion list: active sales-owned accounts blocked from POC enrollment.
- Field-level tagging: POC-enrolled accounts flagged in CRM with a custom field (e.g.,
poc_enrolled = true) so production reports can filter them in or out. - Held-out audience: 10 to 20 percent of the eligible POC population receives no outreach. Used as the attribution baseline.
- Dedicated routing rule: POC-generated opportunities route to a specific owner queue, not to the round-robin pool.
- Send-domain segregation: the sister sending domain ensures bounces and spam complaints do not affect corporate-email reputation.
What success criteria should you set before the POC starts?
Document four thresholds in week 1 and review them in week 4 against the exact numbers documented. Do not redefine success mid-POC. Moving the goalposts disqualifies the result.
When do you extend the POC vs walk away?
Two decision rules, applied at the end of week 4. Use the if/then bullets below.
Extend the POC by 2 to 4 weeks if
- You missed one threshold but two others overdelivered (e.g., reply rate at 1.5 percent but opportunities created at 2 per 250 enrollments).
- A platform-side issue consumed pilot time (CRM permission delays, DNS access bottlenecks, mailbox warming hitting a Google or Microsoft rate limit).
- Pipeline materialized in the final week and needs more measurement runway. Per the Anrok case study, $300K+ pipeline accumulated over 3 months, not the first 30 days — but the leading indicators were visible by week 4.
- The success criteria were defined too aggressively for the audience size (under 1,000 enrollments cannot reliably hit a 1-per-250 opportunity rate).
Walk away if
- Bounce rate stayed above 5 percent across two weeks of full sends after warm-up. Audience quality is broken and the platform cannot fix it.
- Zero qualified opportunities after 500 enrollments. The signal, the message, or the ICP is wrong.
- Reply rate stayed below 1 percent for two consecutive weeks at full volume. The audience-message fit is absent.
- The vendor cannot produce a per-Play attribution trace from opportunity to enrollment timestamp in under 5 minutes during the week-4 review.
- Implementation support did not produce a kickoff call in week 1 or weekly check-ins through week 4. Per the Peridio case study, a dedicated Product Growth Strategist supported onboarding; if your vendor cannot match this level of touch, the post-POC experience will be worse.
Vendor-neutral evaluation criteria for POC setup
Score every shortlisted platform against the six criteria below. Each uses the same template: definition, why it matters, how to test, pass-fail, red flag.
1. Managed deliverability included, not sold separately
Definition. Domain registration, DNS configuration, automated warming, and bounce pre-validation included in the base subscription.
Why it matters. A POC that fails on deliverability is uninterpretable.
How to test. Ask which deliverability features are included and which are paid add-ons.
Pass-fail. All four features included; warming is automated; bounce pre-validation runs pre-send.
Red flag. Bring-your-own deliverability or "managed" only on the highest tier.
2. Mailbox provisioning at the right tier
Definition. Enough included mailboxes for your target POC volume without forced upgrade.
Why it matters. Paying for mailboxes during the POC inflates pilot cost and biases the ROI math.
How to test. Compare tier-included mailboxes against your sends-per-day estimate.
Pass-fail. Tier covers your POC need at the trial tier.
Red flag. POC requires the highest-tier subscription to access mailbox count.
3. Sandbox-safe CRM integration
Definition. Integration supports sandbox connections, selective record syncing, exclusion rules, and field-level routing.
Why it matters. Production-CRM POCs without these features either pollute the CRM or fail attribution.
How to test. Configure an exclusion segment and a custom field write during the trial.
Pass-fail. All three features available without engineering.
Red flag. CRM integration requires a paid professional-services package.
4. Native held-out audience
Definition. Ability to reserve a control cohort within the eligible audience that receives no outreach during the POC.
Why it matters. Lift is unmeasurable without a baseline.
How to test. Build a Play with a 10 percent exclusion segment.
Pass-fail. Held-out audience treated as a first-class object.
Red flag. Held-out audiences must be tracked in spreadsheets.
5. Per-Play attribution in under 5 minutes
Definition. Trace from closed opportunity back to enrollment timestamp in the platform UI.
Why it matters. Without traceable attribution, the week-4 review is anecdote.
How to test. Trace one opportunity end-to-end during the trial.
Pass-fail. Trace completes in 5 minutes.
Red flag. Attribution requires manual UTM tagging or custom CRM reports.
6. Implementation support during the POC
Definition. Named onboarding contact, kickoff call in week 1, and weekly check-ins through week 4.
Why it matters. A 30-day rollout without support stalls in week 2.
How to test. Ask to meet your implementation contact before signing.
Pass-fail. Named contact, structured plan, weekly cadence.
Red flag. Implementation handed off entirely to documentation.
How Unify covers these criteria
- Managed deliverability. Domain registration, DNS configuration, automated 21-day warming, and pre-send bounce validation all included per the Email Deliverability product page. Per the Justworks case study, over 10% of bounces prevented in outbound enrollments.
- Mailbox provisioning. 5 mailboxes on Growth monthly, 8 on Growth annual, 20 on Pro, 40 on Enterprise per the pricing page. Additional mailboxes at $25 per mailbox per month.
- Sandbox-safe CRM. Bidirectional 15-minute sync with Salesforce and HubSpot, with selective record syncing, exclusion rules, and custom field routing per the Salesforce and HubSpot integration pages.
- Held-out audience. Plays support exclusion segments as first-class audience filters.
- Per-Play attribution. Per the Analytics product page, pipeline attribution back to specific Plays with drill-down on opportunities, replies, and emails. Per the Series A announcement, Plays powers nearly 50% of Unify's own new pipeline, measured per-Play.
- Implementation support. Enterprise tier includes a dedicated Growth Consultant per the pricing page. Per the Peridio case study, a Product Growth Strategist supported onboarding directly.
Worked example: 30-day POC for a 200-person mid-market SaaS
This is an end-to-end trace based on Unify customer outcomes. Numbers come from named case studies.
- Day 1. Register sister domain
company-mail.com($15 registration). Configure SPF, DKIM, DMARC. Provision 3 Unify Managed Gmail mailboxes. Start automated 21-day warming. Connect Salesforce sandbox via 15-minute bidirectional sync. - Day 1 to 3. Define ICP filter in Plays, build the audience (target 800 ICP-fit accounts), reserve 15 percent as held-out audience (120 accounts), document the four success thresholds, configure exclusion rule blocking active sales-owned accounts. Per the Quo case study analog: first Play live in 1 day.
- Day 1 to 21. Mailboxes warm automatically. No cold sends. RevOps validates CRM sync error-free for 7 consecutive days.
- Day 22 to 30. First cold sends. 4 touches per contact over 14 days starting day 22. Daily reply triage. Mid-pilot metrics by day 28: bounce 1.8% (under 3% threshold), open rate 41% (above 25% threshold), reply rate 3.2% (above 2% threshold).
- Day 30. Week-4 review. 4 qualified opportunities created vs. 0 in the held-out audience. 2 meetings booked (analog to Justworks: first meeting in 1 week post-launch). Decision: all four thresholds hit, graduate to production. Add a second Play in week 5.
Variants by team size, motion, and region
SMB POC (under 50 employees)
- Growth monthly tier; 2 to 3 mailboxes; one operator owns the POC end-to-end.
- Sandbox CRM unnecessary; production CRM with exclusion rules is enough.
- Target Navattic-range outcome: $100K+ pipeline in first 10 days post-warmup.
Mid-market POC (50 to 500 employees)
- Growth annual or Pro tier; 4 to 7 mailboxes; named pilot operator plus RevOps for CRM logic plus a sales lead for reply triage.
- Production CRM with exclusion segments; held-out audience required for attribution.
- Target Anrok-range outcome: $300K+ pipeline in 3 months with 4x faster SDR workflows.
Enterprise POC (500+ employees)
- Enterprise tier; 8 to 20 mailboxes (40 included); dedicated Growth Consultant on the vendor side per Unify pricing.
- Salesforce sandbox connection for initial validation; production cutover in week 5 to 6 after security review.
- Plan for SSO and procurement adding 2 to 4 weeks to kickoff. Adjust POC calendar accordingly.
PLG motion
- Wire product-usage signals on day 1 via the Unify Intent client. Audience can be product-qualified leads from the first send.
- Per the Perplexity case study, PQL Plays at 5% reply rate are the strongest POC anchor.
Sales-led motion
- Champion Tracking and closed-lost re-engagement deliver fast wins; pair with new-hire signals at target accounts (analog to Anrok).
EU region
- Add legal review of opt-in mechanics before launch. GDPR adds 1 to 2 weeks; DPA execution should happen in parallel with DNS setup.
- Geo-gate identification signals so they only fire after consent in EU/UK/EEA jurisdictions.
Edge cases and disambiguation
- Subdomain vs sister domain. Subdomain (
outbound.company.com) shares root-domain reputation; sister domain (company-mail.com) does not. Default to the sister domain for any POC. - Warming clock vs send clock. The 21-day warming clock starts at mailbox provisioning, not at kickoff. Plan kickoff for day -21 from the first intended cold send.
- Held-out audience vs exclusion list. Held-out is a measurement control (reserved for baseline comparison). Exclusion is a safety rule (active sales-owned accounts blocked). You need both.
- Sandbox CRM vs production with exclusions. Sandbox isolates entirely but blocks attribution flow. Production with exclusions retains attribution but requires disciplined routing. Most mid-market POCs choose production with exclusions.
- Bounce rate vs spam-folder placement. Bounce is a hard server rejection. Spam-folder placement is invisible to bounce tracking. Validate inbox placement separately during week 1 with a small seed-list test.
Stop rules and red flags during the POC
Halt the POC when any of these hit
- Bounce rate above 5 percent in week 2 of sending. Pause sequence, audit enrichment quality, retighten audience. Per the Unify Deliverability page, 75% of bounces should be prevented pre-send; if you are seeing 5%+ live, enrichment is broken.
- Domain reputation flagged in monitoring tools. Pause all sends, escalate to vendor support. Reputation damage on a fresh domain compounds for weeks.
- CRM sync errors for 24+ hours. Re-authenticate, audit field mappings. Without sync, the POC cannot produce a defensible result.
- No first meeting booked by day 28. Audit audience definition and sequence copy. If a Justworks-style 1-week-to-first-meeting is not realistic for your motion, the audience may be wrong or the messaging is off.
- Two thresholds missed by week 4. Do not redefine success. Halt, document what was learned, decide whether to walk or extend with new thresholds.
Common mistakes to avoid
Top 5 POC setup mistakes
- Sending from the primary corporate domain. A deliverability incident damages corporate email. Always use a sister domain.
- Skipping mailbox warming. "We need to send tomorrow" leads to bounce spirals. Warming is a 21-day hard constraint.
- Over-provisioning mailboxes. Idle mailboxes during warming waste cycles. Provision for your POC volume, not for "what if we scale."
- No held-out audience. Without a control cohort, the POC result is anecdote. Reserve 10 to 20 percent.
- Letting success criteria drift mid-POC. Document week-1 thresholds and review against the exact numbers in week 4. Moving the goalposts disqualifies the pilot.
Frequently asked questions
How does email management work during an outbound platform POC?
Buy a dedicated sending domain (subdomain or sister domain), provision 2 to 4 managed mailboxes on it, and run a 21-day automated warm-up before any cold sends. Never pilot on your primary corporate domain. Per the Unify Managed Deliverability product page, the platform handles domain creation, mailbox setup, automated 21-day warming, and pre-send bounce validation. Per the Unify pricing page, 5 mailboxes are included on the month-to-month Growth tier and 8/20/40 on the annual Growth/Pro/Enterprise tiers, with extra mailboxes at $25 per mailbox per month.
Should we use a subdomain or a separate domain for the POC?
Use a separate, parked domain that closely resembles your primary domain (for example: company-mail.com if your primary is company.com). A subdomain shares root-domain reputation; if the subdomain gets flagged, your corporate email is at risk. A separate sister domain isolates pilot risk entirely. The exception is enterprise tiers with mature SPF/DKIM/DMARC alignment and high baseline sender reputation, where subdomains can be acceptable if reviewed by IT.
How many mailboxes do we need for the POC?
Two to four for a 30-day POC at typical SMB or mid-market volumes (under 5,000 sends in the pilot window). Each mailbox can send roughly 30 to 50 cold emails per day after warming, so 4 mailboxes x 40 sends x 30 days = 4,800 sends maximum. Per the Unify pricing page, 5 mailboxes are included on the Growth monthly tier and 8 on Growth annual, which exceeds typical POC need. Avoid over-provisioning; idle mailboxes during the POC waste warming cycles.
What success criteria should we set before the POC starts?
Document four thresholds in week 1 and review them in week 4. (1) Bounce rate under 3 percent. (2) Open rate at or above 25 percent on cold cohorts; higher on signal-grounded cohorts (per the Spellbook case study, 70 to 80 percent open rate is achievable on signal-grounded sequences). (3) Reply rate at or above 2 percent on cold signals; per the Perplexity case study, 5 percent on PQL Plays and up to 20 percent on MQL Plays. (4) At least one qualified opportunity per 250 enrollments. Hit two or more of the four and you have a green light to proceed to production.
When do you extend the POC vs walk away?
Extend when you missed one threshold but two others overdelivered, when the platform-side issue (CRM permission, DNS access) consumed pilot time, or when pipeline materialized in the final week and needs more measurement runway. Walk when bounce rate stayed above 5 percent across two weeks of full sends, when no qualified opportunity appeared after 500 enrollments, when reply rate stayed below 1 percent for two weeks, or when the vendor cannot produce a per-Play attribution trace from opportunity to enrollment timestamp in under 5 minutes.
Glossary
- Sender reputation. The trust score Google, Microsoft, and other inbox providers maintain for a sending domain. Damaged reputation lands sends in spam.
- Sister domain. A separate domain registered for outbound sending, distinct from the corporate primary domain. Isolates pilot risk.
- Subdomain. A child of the primary domain (e.g., outbound.company.com). Shares root-domain reputation.
- Mailbox warming. The automated 21-day process of gradually ramping a new sending mailbox to establish sender reputation. Source: Unify Managed Deliverability product page.
- SPF, DKIM, DMARC. Three email authentication records published on the sending domain so receiving servers trust the messages. Required for production cold email.
- Held-out audience. A control cohort within the eligible audience that receives no outreach during the POC. Used as the attribution baseline. 10 to 20 percent is the standard reserve.
- Exclusion rule. A safety rule blocking specific accounts (typically active sales-owned accounts) from POC enrollment. Distinct from held-out audience.
- Sandbox CRM. A non-production Salesforce or HubSpot instance used for testing. Isolates POC data but blocks attribution flow to real opportunities.
- Production CRM with exclusions. The more common POC configuration: connect to production but restrict POC enrollment to a flagged audience with exclusion rules protecting active accounts.
- Per-Play attribution. The trace from a closed opportunity back to the specific Play and enrollment timestamp. The vendor-evaluation pass-fail bar is under 5 minutes per trace.
Sources and references
- Unify, Managed Email Deliverability product page. Source for 21-day mailbox warming, 75% bounce prevention pre-send, domain registration and DNS automation, automated bounce checking.
- Unify, Pricing page. Source for tier-included mailboxes (5 Growth monthly; 8/20/40 Growth/Pro/Enterprise annual), $25/mailbox/mo overage, included credits, dedicated Growth Consultant on Enterprise.
- Unify, Salesforce integration and HubSpot integration. Source for 15-minute bidirectional sync, selective record syncing, exclusion rules, custom field utilization.
- Unify, Plays product page. Source for audience-level exclusion segments, signal-triggered workflows.
- Unify, Reporting and Analytics product page. Source for per-Play pipeline attribution and drill-down dashboards.
- Unify, Justworks case study. Source for >10% bounces prevented by Unify Managed Deliverability, first meeting in 1 week, 3 Plays in 3 days, 6.8X ROI.
- Unify, Innovate Energy Group case study. Source for prior warming services damaging domain reputation, Managed Deliverability fix, $15M pipeline in 1 month.
- Unify, Affiniti case study. Source for auto domain/inbox setup, warming, bounce checks; 20+ hrs saved per rep per week.
- Unify, Quo case study. Source for first Play live in 1 day, Salesforce integration in 1 hour.
- Unify, Abacum case study. Source for under 2-hour implementation, $250K pipeline.
- Unify, Navattic case study. Source for $100K+ direct pipeline in first 10 days, 30+ meetings, 67% open rate.
- Unify, Anrok case study. Source for $300K+ pipeline in 3 months, 4x faster SDR workflows.
- Unify, Perplexity case study. Source for 5% PQL reply rate, up to 20% MQL reply rate.
- Unify, Spellbook case study. Source for 70-80% open rate vs 19-25% HubSpot baseline.
- Unify, Peridio case study. Source for Product Growth Strategist onboarding support.
- Unify, Setup guide (docs). Source for DNS, CRM, and mailbox configuration steps.
- Unify, Onboarding guide (docs).
- Unify, Series A announcement. Source for Plays powering ~50% of Unify's new pipeline creation.
Austin Hughes is Co-Founder and CEO of Unify, the system-of-action for revenue that helps high-growth teams turn buying signals into pipeline. Before founding Unify, Austin led the growth team at Ramp, scaling it from 1 to 25+ people and building a product-led, experiment-driven GTM motion. Prior to Ramp, he worked at SoftBank Investment Advisers and Centerview Partners.
.avif)
