Marketing Outbound Reach Forecasts: Test Who Is Matched, Eligible, and Contactable

An outbound reach estimate is meaningful only when its population, unit, exclusions, and evaluation time are clear. Before treating a displayed count as a launchable audience, trace it through matched accounts, selected people, available contact routes, applicable restrictions, and actual enrollment results.

A count can be correct for one stage and misleading for another. Matching companies are not the same as unique people. People with a known email address are not necessarily eligible for outreach. Eligible enrollments are not guaranteed sends, replies, or pipeline.

This guide provides a proposed acceptance worksheet for evaluating campaign reach. It does not assert an observed forecast-accuracy benchmark or undocumented semantics for a particular Unify reach display.

Define what the estimate counts

Ask the vendor to identify the unit beside every displayed number. Is it companies, people, records, email addresses, selected rows, or potential enrollments? If the interface does not state the unit, make that an explicit question before discussing launch volume.

Record the query, filters, source population, and time of the estimate. Include whether the result is a saved set or a changing audience. Unify's list-building overview distinguishes lists whose membership is controlled through actions from audiences whose results change with saved conditions.

That distinction matters when a campaign review and launch occur at different times. A changing audience can legitimately include different records later. A list's displayed table can also be narrowed by search or filters without changing its underlying membership. Preserve the exact state used to make the decision.

Use the audience creation guide to inspect the saved selection criteria. The goal is to explain the count in the product's actual terms, not to relabel every preview as “people we can email.”

Reconcile the stages instead of multiplying assumptions

Build a stage-by-stage account of the campaign population. Keep each stage's unit and exclusion basis visible. A vendor should be able to explain which stage a forecast represents and what remains to happen afterward.

Do not assume every account contributes the same number of people or every person contributes one usable route. Do not add account counts and contact counts together. If the forecast changes units between screens, document the transition explicitly.

The find and enrich target account contacts motion provides useful context for the account-to-person transition. The company and contact data overview describes the data layer. Neither guarantees that every matched account will produce an eligible recipient for a specific campaign.

Inspect the selection logic behind the forecast

Review company criteria and contact criteria separately. A company can meet the market definition while the selected person does not hold the role needed for the campaign. A contact can have the right title while working at a different entity than the one the campaign intends to target.

Use filter builder to inspect grouped conditions and related-record logic. The difference between requiring all criteria and allowing any criterion can materially change a population. The evaluator should be able to explain the saved logic without relying on the preview's size as proof of correctness.

If research qualifies records before prospecting, identify that stage and its output. Play actions can perform research, prospecting, and enrollment as separate operations. A count before qualification should not be presented as the count after qualification unless the product has actually evaluated those records.

Record any per-account contact limit, persona rule, or unresolved research condition that affects delivery of the selected people. Treat a missing value as unresolved where appropriate rather than automatically including or excluding it without an explicit rule.

Check contactability without equating it with permission

A stored email address establishes that an address is present. It does not establish that the person may be contacted, that the address is current, or that a particular message will be delivered. Similar distinctions apply to phone numbers and calling eligibility.

Ask what verification or enrichment has occurred, when it occurred, and whether the estimate depends on work that has not run yet. Keep unavailable routes separate from routes that are present but restricted. Both can reduce launchable reach, but they require different decisions.

When building lists with chat, inspect missing values and supporting evidence before relying on the result. The documentation notes that companies need usable website domains and people need email addresses to be saved from new prospect data. A failed save or incomplete source record should not silently disappear from the forecast reconciliation.

Avoid treating contactability as an invitation to bypass opt-outs or account protections. The reach review should explain why a record cannot proceed, so the team can distinguish a data problem from a legitimate restriction.

Make exclusions and ownership rules explicit

List every exclusion that applies to the campaign: existing customers, open opportunities, competitors, internal accounts, opted-out people, or other team-defined groups. Record whether each restriction is evaluated at preview, list creation, enrollment, later execution, or more than one point.

Unify's sequence exclusions match records as data changes. The documentation explains that exclusions can affect outreach after enrollment as well. That means a preview is not a permanent eligibility certificate for every record it contains.

For users assigned to engagement policies, policy exclusions replace sequence ruleset exclusions. The evaluator must identify which users and policies govern the planned action. Counting under one team's rules and launching under another team's identity can produce a different eligible population.

Inspect account eligibility for claiming and engagement restrictions. Account ownership and outreach permission are separate. A record can match the audience and still require a claim, show a warning, or be blocked.

Do not sum overlapping exclusion counts as though they describe distinct people. One person may match several restrictions. For a reconciled total, retain unique IDs and either use a defined primary exclusion reason or calculate the union of excluded records. Preserve secondary reasons without double-counting them.

Distinguish enrollment eligibility from actual enrollment

A selection screen shows what the operator selected under its current search and filters. The eventual result should show what the system actually accepted. Unify's Lists documentation describes reviewing selected counts, eligibility, and existing-enrollment conflicts before confirmation, then using Activity to inspect enrolled, skipped, and failed records.

The same documentation states that up to 5,000 people can be enrolled at a time from a list. That is an operation limit, not a promise of campaign reach or a cap on every possible data surface. If a campaign needs several operations, preserve the batch boundaries and unique people so retries do not inflate the reported total.

Sequence enrollments track each person's status, mailbox, completed steps, and upcoming work. Use these records to distinguish successful enrollment from queued, blocked, excluded, canceled, or other outcomes.

Existing simultaneous-enrollment and re-enrollment settings also affect the result. Review the applicable sequence ruleset instead of assuming every selected person can enter every sequence. A deliberate conflict restriction should be reported as such, not classified as unexplained forecast failure.

For agentic content, the entry path can affect whether a draft-review step occurs. Document the chosen path and any remaining approval requirement. An estimate should not be described as immediate sending volume when work still needs review.

Separate reach from when messages can send

Even an eligible enrollment may wait for earlier steps, delays, a permitted send window, mailbox readiness, or shared capacity. Unify's sending limits explain that sequences sharing a mailbox share its available capacity, and follow-ups use capacity alongside first touches.

The send schedule determines allowed windows. It does not make every enrolled person immediately send-ready. A launch forecast that names a day or week should state how schedule and capacity constraints were incorporated.

Keep this check focused on interpreting the estimate. The reach worksheet is not a staffing plan for handling replies or a recommendation to raise mailbox limits. A forecast can be reconciled correctly while still describing work that will take time to execute.

The sequencing overview explains the broader outreach workflow. The evidence for launch timing should come from the configured sequence, mailbox, and enrollment states rather than the product category alone.

Use a reach-estimate acceptance worksheet

This blank worksheet can be completed during a vendor review or an authorized campaign preparation. It contains no observed forecast or enrollment result.

Attach the saved query or audience reference, selected user and policy, sequence version, and any operation limit. Keep a record of the specific IDs used for reconciliation in an approved internal system. Do not publish prospect information as part of the evaluation.

When the preview and enrollment differ, classify the difference. Useful explanations include changed source data, different user eligibility, active-enrollment conflicts, missing contact routes, operation limits, failed processing, or an unresolved discrepancy. Do not force every difference into a targeting-quality category.

Ask the vendor to explain any remaining count that cannot be traced to records or a documented rule. An estimate can be useful with explicit limitations. An unexplained count should not be represented as a guaranteed audience merely because it appears on a dashboard.

Define quantitative checks with the same population

Choose a fixed set of unique selected people at the preview cutoff. Record the campaign definition, time, and exclusions known at that point. For enrollment realization, use the number of those same people with a confirmed enrollment by the later cutoff as the numerator and the original selected population as the denominator.

Report skipped, failed, still-processing, no-longer-eligible, and unresolved records separately. Do not remove them from the denominator after the fact to make the realization rate look higher. If the original denominator is zero, report not applicable.

If the audience changes between the two cutoffs, identify added and removed records separately. A count difference alone cannot reveal whether the same people were processed. Use stable IDs and the stated snapshots to avoid comparing two different populations as though they were one cohort.

A reach forecast is not a reply forecast. Sequence results have their own outcome populations and denominators. Do not multiply a reach count by an unrelated reply-rate card and present the result as observed pipeline potential.

Make the launch decision from resolved states

Approve the campaign when the team understands what was counted, which records can proceed, what is still pending, and how the result will be checked. Record any limitation that affects the planned launch window or audience composition.

If a mandatory exclusion is unexplained, resolve it before outreach. If a data route is missing, decide whether to enrich, narrow the campaign, or leave the record unresolved. If the difference is simply a documented operation limit, plan bounded batches and reconcile them by unique record identity.

The strongest vendor demonstration is an understandable path from a defined audience to actual eligible enrollment, with the exceptions visible. It does not require a promise that every matched account becomes a send, a reply, or an opportunity.

Ask the vendor to explain a changed count

A useful demonstration includes an explanation of a count transition, not just the final number. Ask which records moved out of the defined population and why. The answer should distinguish a change in source data from a changed filter, an eligibility restriction, a processing failure, or an operation limit.

Request record-level evidence where the product provides it. If only aggregate counts are available, state the resulting limitation: the evaluator may understand the total difference without being able to prove which people it represents. Do not describe that weaker evidence as full identity-level reconciliation.

Keep the same user and configuration when the purpose is to compare preview with enrollment. If the demonstration switches to an administrator or a different sender, record the change. The resulting count may be valid for that new setup while failing to answer the question about the intended campaign operator.

Preserve decision boundaries when the campaign changes

A revised audience definition creates a new basis for the forecast. Give the new definition an identifier or dated reference and retain the earlier review. Do not overwrite the original estimate and then present a later enrollment as if it had been predicted by the revised number all along.

The same applies when the team changes exclusions, persona selection, or sender assignment. Record which population was reviewed under which rules. A later comparison should separate the original cohort from newly added people and people removed before execution.

This does not require freezing every business process indefinitely. It requires enough history to explain the decision. The team can keep working with current data while retaining a clear internal record of the estimate that supported the launch and the changes made afterward.

Decide which uncertainty is acceptable

Some uncertainty is ordinary in a changing campaign population. The buying decision should distinguish a bounded, explained difference from an unresolved control requirement. A known delay in enrichment may be acceptable for a later launch; an unexplained customer exclusion may block the same launch.

Name the owner and resolution path for each unresolved category. A data owner may need to inspect contact routes, an administrator may need to review a policy, and the campaign owner may need to adjust the intended launch window. A single generic backlog item makes those different responsibilities harder to manage.

Keep the approval conditional where necessary. State which population is ready, which records remain outside the approved launch, and what evidence would allow those records to proceed. That is more useful than forcing the entire forecast into a binary accurate-or-inaccurate label.

The final acceptance record should let another operator explain the displayed number without guessing: what it counts, when it was evaluated, which rules it includes, and what still has to happen before outreach. If those answers are clear, the estimate can support planning even when it is not a guarantee of final sends.

Frequently asked questions

Is a matched-account count the number of people we can email?

No. Accounts, selected people, contact routes, eligible people, and actual enrollments are different stages. Establish the unit and restrictions before interpreting a forecast.

Why can eligibility change after the preview?

Source data, exclusions, ownership rules, and existing enrollment state can change. Record when each check occurs and reconcile the actual result against the original population.

Can exclusion counts be added together?

Only if they are defined as non-overlapping categories. Otherwise, one record can match several exclusions. Use unique IDs or a defined primary reason to avoid double-counting.

Does successful enrollment mean an email sends immediately?

No. Steps, delays, send schedules, mailbox capacity, and approval or blocked states can affect execution. Inspect the enrollment and sending configuration separately.