Minimum Outbound Routine: Respond, Advance, Open

Minimum means complete, not busy

A solo founder does not need a miniature sales department. The routine only needs to preserve the complete path from evidence to conversation to next action.

Salesforce describes B2B prospecting as research, prioritisation, contact, nurturing and qualification (B2B prospecting guide). HubSpot also distinguishes inbound and outbound prospecting while keeping research and qualification inside the discipline (sales prospecting guide). A routine that sends new messages but ignores replies and qualification is therefore incomplete.

The minimum useful system has three queues: respond, advance and open.

Queue 1: respond to buyers

Read new replies before starting new outreach. Classify each one by the action it requires:

• answer a factual question;
• propose a next step;
• clarify fit;
• respect a refusal or contact preference;
• record that no action is required.

Do not hide buyer responses inside an inbox. Every meaningful reply needs a dated next action, a closure reason or an explicit pause condition. This queue comes first because the buyer has already spent attention on the conversation.

Queue 2: advance existing conversations

Review open conversations and active opportunities. For each one, write the last verified fact and the next decision required.

Examples:

• The buyer asked for proof. Send one relevant case or explain that the proof is not available.
• The contact is interested but not the decision maker. Ask how the decision is made instead of guessing the org chart.
• The timing is wrong. Record the event or date that would justify reopening.
• Fit is weak. Close the record with the reason rather than adding another reminder.

An opportunity without a next action is not active. It is an unresolved record consuming attention.

Queue 3: open new accounts

Only create new conversations after the first two queues are under control. Each account must have four fields before contact:

1. the ICP reason;
2. the current evidence or signal;
3. the relevant person or contact path;
4. the question or useful observation that justifies the message.

If one field is missing, keep the account in research or reject it. Do not turn missing evidence into generic personalisation.

The weekly loop

Start of week: choose and prepare

Select a small working list that fits the available capacity. Review every active conversation first, then prepare new accounts. The list is complete when each account has a reason to contact, not when it reaches a round number.

During the week: run the three queues in order

At every outbound block:

1. handle meaningful replies;
2. complete promised next actions;
3. open new conversations from the prepared list.

This ordering prevents new activity from displacing work already earned.

End of week: close the loop

Review every record touched during the week. Close, pause or date the next action. Record repeated objections, invalid targeting assumptions and signals that produced a useful conversation.

Then choose one process change for the following week. Changing several rules at once makes the result impossible to interpret.

A definition of done for every account

An account leaves the weekly list in one of five states:

• next action dated: a person owns a specific continuation;
• waiting with a condition: a known event or date triggers review;
• not now: timing is wrong and the reason is recorded;
• not a fit: evidence contradicts the ICP;
• do not contact: a preference or objection blocks outreach.

"Follow up later" is not a state. It lacks an owner, condition and date.

What the minimum routine measures

Track only the numbers that help change a decision:

• prepared accounts that passed the evidence check;
• meaningful buyer replies handled on time;
• active conversations with a dated next action;
• conversations closed for poor fit;
• repeated objections;
• new accounts left uncontacted because evidence was weak.

The last measure is useful. A disciplined rejection protects time and reputation. Raw sends and opens may describe activity, but they do not prove fit or progress.

Where Lead Intelligence fits

Lead Intelligence can support the preparation and priority work after the founder has defined an ICP and offer. Ember describes it as using context and signals to surface opportunities and recommend a next action (official Lead Intelligence page).

Use it to explain why an account enters the working list. Keep the human decision at the boundary: a signal is a reason to investigate, not proof of buying intent. Contact preferences, refusals and legal requirements still govern the action.

Use the qualification framework to define fit. Use the weekly outbound review guide when the routine needs a deeper team review.

Stop rules

Pause new outreach when:

• buyer replies are waiting;
• promised actions are late;
• active records have no clear state;
• prepared accounts lack evidence;
• objections or do-not-contact requests are not applied;
• the founder cannot complete the weekly closure review.

The remedy is to restore the three queues, not to add automation.

Limits and sources

This routine does not prescribe a universal number of accounts, messages or hours. Capacity varies with deal complexity, channel, buyer expectations and the founder's other work. The three-queue order is an Ember editorial method for maintaining a complete process.

Sources reviewed on August 3, 2026: Salesforce B2B prospecting, HubSpot sales prospecting and Ember Lead Intelligence.