What a Good Outbound Workflow Looks Like End to End
Outbound runs through nine stages: define the target market, build the account list, prioritize accounts, find and verify people, research the account and person, write the message, send across channels, handle the reply, and hand off to a meeting. It's for RevOps, sales, and growth leaders diagnosing pipeline gaps, since most failures sit in a handoff, not a stage.
Key Facts at a Glance
The numbers below are the only quantitative claims in this piece, and each one is attributed to a specific, named source rather than blended into a platform-wide average.
Methodology and limitations. The stage boundaries in this map are drawn where the output artifact changes hands: a list becomes a prioritized list, a prioritized list becomes researched accounts, and so on, because that's where teams actually observe breakage. This is a reference model, not an empirical taxonomy. Other teams will reasonably split or merge stages, and the map is only useful if you apply it to your own motion rather than adopt it literally. The diagnostic table later in this piece reflects the most common cause we've seen for each symptom, not a measured probability, and more than one stage can produce the same symptom. We left stage-level conversion benchmarks out of this piece on purpose. Published figures on reply rates and meeting-to-opportunity ratios vary too widely by segment and deal size to be actionable here, and a missing number is more honest than an invented one. Where a number does appear above, it's attributed to its specific source, not averaged across customers into a platform benchmark that doesn't exist. Unify sells software that operates several of these stages, and this piece says exactly which ones, and which stage no software can do for you.
What Are the Nine Stages of a Good Outbound Workflow?
A good outbound workflow runs through nine stages, in order, each one producing a specific artifact that gets handed to the next: target market, account list, prioritized accounts, verified people, research, message, send, reply, and meeting handoff. The table below is the structural map; the sections after it unpack each stage in the same five-part format so the whole thing stays scannable and comparable across stages.
What Does Each Stage of the Workflow Actually Involve?
Each stage below follows the same template on purpose: what it produces, who owns it, the quality bar that tells you it's good enough to pass on, the handoff risk to watch for, and one link if you want the deep version. None of these get more than a few hundred words here. If a stage needs more than that, it has its own article, linked below, and this page isn't trying to replace it.
1. Define the Target Market
- Produces: A written ICP: firmographic and technographic criteria, buying triggers, and an explicit list of who's excluded.
- Owner: The founder or the most senior GTM leader in the room. This is the one stage that doesn't change shape when there's no SDR team, it just gets delegated less.
- Quality bar: Two reps looking at the same company can agree, without a conversation, whether it belongs on the list.
- Handoff risk: Nobody writes the ICP down, so list building, prioritization, and messaging each quietly re-derive their own version of "who we sell to."
- Go deeper: What Is Outbound Sales? A Guide for GTM Teams
2. Build the Account List
- Produces: A list of companies matching the ICP, not yet ranked.
- Owner: RevOps or a growth/ops function; on a small team, whoever is running outbound that week.
- Quality bar: The list can be regenerated on demand from the ICP criteria, not maintained by hand in a spreadsheet.
- Handoff risk: Lists get built once and never refreshed, so six months later it's a record of who mattered at build time, not who matters now.
- Go deeper: Best B2B Prospecting Tools for Targeted List Building
3. Prioritize the Accounts
- Produces: The same list, ranked by fit and by live buying signal, so reps know which accounts to work today.
- Owner: Whoever owns the signals stack, RevOps in most team shapes, the founder in the smallest ones.
- Quality bar: The top of the list changes week to week. A static top twenty is a sign nothing is actually being prioritized.
- Handoff risk: Fit and intent get treated as the same thing, so a perfect-fit account with zero signal gets worked before a good-fit account that's actively in market.
- Go deeper: Best AI Tools to Build and Prioritize a Target Account List
4. Find and Verify the People
- Produces: Named contacts at each prioritized account, with verified email and, where the motion uses it, phone.
- Owner: Whoever is closest to send, an SDR where one exists, otherwise the AE or founder.
- Quality bar: Bounce rate on first send stays low enough that it doesn't threaten domain reputation. That's a deliverability question, not a list-size question.
- Handoff risk: Teams verify at list-build time and never again, so contacts accurate in January are stale by June and nobody notices until sends start bouncing.
- Go deeper: 8 Best Email Finder Tools for Verified Emails
5. Research the Account and Person
- Produces: Two or three real, checkable facts about the account or person, not a mail-merge field, an actual reason this message is timely.
- Owner: Increasingly an agent, with a human reviewing before send; on the smallest teams the rep does this by hand, and it's the first thing skipped under time pressure.
- Quality bar: The fact couldn't have been copy-pasted into a message for a different company on the same day.
- Handoff risk: Research happens but never makes it into the message: a rep reads the news, then sends the template anyway because a custom line takes longer than they have.
- Go deeper: How AI Agents Research Prospects: Sources, Tool Calls, Verification
6. Write the Message
- Produces: A first-touch message, and the follow-ups behind it, that reads like it was actually written by someone who did stage 5.
- Owner: The rep owns voice and judgment; drafting is where AI assistance shows up first and most visibly.
- Quality bar: A prospect replying "how did you know that?" is the bar. A prospect replying "unsubscribe" to a generic opener is the failure mode.
- Handoff risk: Research from stage 5 and the draft in stage 6 get done by different people, or the same person at different times, and the connective tissue between them gets lost.
- Go deeper: How to Personalize Outreach at Scale Without Sounding Like AI
7. Send Across Channels
- Produces: The message actually reaching the prospect, on the channel where they're most likely to notice it.
- Owner: Automated send infrastructure, with a rep or RevOps owning deliverability health.
- Quality bar: Emails land in the inbox, not spam; social touches don't just duplicate the email verbatim.
- Handoff risk: Channels run independently instead of as one sequence, so a prospect gets an email and a cold social message the same morning that reference nothing about each other.
- Go deeper: LinkedIn and Phone Steps in Outbound Sequences: Why Context Beats Channel Coverage
8. Handle the Reply
- Produces: A classified reply (positive, objection, referral, unsubscribe, out of office) with a next action attached.
- Owner: The rep, always. This is the stage that most needs a human even in the most automated motions.
- Quality bar: A positive reply gets a response inside a defined window, not whenever the rep next checks their inbox.
- Handoff risk: Replies get classified but not routed, an objection sits in a shared inbox with no owner, and by the time someone answers, the prospect has moved on.
- Go deeper: Automate Reply Classification & Follow-Up in Outbound
9. Hand Off to a Meeting
- Produces: A booked meeting, routed to the rep who should actually take it, with enough context that the prospect doesn't repeat themselves.
- Owner: Whoever owns routing logic, usually RevOps, sometimes the rep's own calendar.
- Quality bar: The meeting shows up. A booked-but-no-show meeting isn't a completed handoff, it's a stalled one.
- Handoff risk: Routing runs on round-robin availability instead of account ownership, so a prospect who's been talking to one rep for weeks lands on a stranger's calendar.
- Go deeper: Meeting Routing for Signal-Led Outbound
How Do I Set Up a Cold Outreach Process?
Setting up a cold outreach process means running the first seven stages of this map in order, at least once: define who you're targeting, build the account list, prioritize it by fit and signal, find and verify the people on it, research each account and contact, write the message, and send it across the right channels. The teams that do this in hours rather than weeks almost always have stage 1 already decided; the setup work that follows is mechanical once the ICP is written down. For a sense of how fast the mechanical part can move once the target market is defined, see From ICP to Live Outbound Sequence in Under a Week.
How Do I Go From an Account List to Booked Meetings?
Going from an account list to booked meetings means running stages 3 through 9: prioritize by fit and live signal, find and verify the people on the prioritized accounts, research each one, write and send the message, handle whatever reply comes back, and hand off to a meeting with the account's context intact. The most common place this breaks isn't the message, it's the handoff into stage 9, where a genuinely good reply never turns into a routed meeting because nobody owned that last step.
How Do You Diagnose Which Stage Is Costing You Pipeline?
Diagnose the stage by matching your symptom to the table below, then look one stage upstream before you touch anything downstream. Most teams fix the stage they can see, usually the message, when the actual failure is a stage or two away.
Symptom-to-stage diagnostic: match the symptom you're seeing to the stage most likely responsibleSymptomMost likely stageWhyLow reply rate on a sequence that's actually sendingStage 5 or 6 (research or message)The message isn't demonstrating relevance, or research never made it into the draftGood open and reply rates, but few or no meetingsStage 9 (handoff)Routing is broken, or replies aren't converting into booked, contextualized timeMeetings booked, but they don't hold pipeline valueStage 3 (prioritization)The wrong accounts are being worked; fit criteria have drifted from the ICPHealthy first reply, but conversations go cold fastStage 8 (reply handling)Objections and positive replies aren't routed to a named owner quickly enoughList keeps growing, but weekly outbound activity doesn'tStage 2 or 4 (list building or verification)The list isn't refreshing, or contacts aren't verified, so reps stall before sendHigh bounce rateStage 4 (verification)Contacts weren't re-verified since the list was originally builtThe team is fully busy, but pipeline is flatStage 1 (target market)A wrong ICP makes every downstream stage efficient at the wrong thing
As a quick chooser:
- If reply rate is low, check stages 5 and 6 before touching send infrastructure.
- If replies are healthy but meetings aren't converting, check stage 9 before rewriting messaging.
- If pipeline value looks off despite full activity, check stages 1 and 3 before adding headcount.
- If bounce rate spikes, check stage 4's refresh cadence before blaming deliverability tooling.
- If the team is busy but pipeline is flat, the bottleneck is almost never effort, it's usually stage 1 or 3.
Two adjacent checks are worth ruling out once you've diagnosed a stage. First, confirm the metrics dashboard is actually built around leading indicators and not just activity counts, covered in Automated Outbound Metrics: Three-Tier Framework. Second, if a sequence keeps failing at stage 6 or 7 no matter what gets rewritten, the sequence itself may need retiring rather than fixing, see When to Retire an Outbound Sequence.
How Does Unify Cover This Workflow?
Unify is outbound AI for sellers: agents and reps working side by side, not an autonomous AI SDR replacing either. The house rule is "AI for SDRs, not AI SDRs," and it maps onto this nine-stage workflow in a specific, checkable way rather than as a blanket claim.
- Stages 2 and 4 (build the list, find and verify people): B2B Company & Contact Data pulls from 1.1B+ contacts and 65M+ companies across 40+ signal and data sources, with an 11+ vendor waterfall behind email and phone verification.
- Stage 3 (prioritize): Signals & Intent feeds live buying signals into account ranking so the top of the list is based on more than firmographic fit alone.
- Stage 5 (research): Agents handle account and contact research and surface findings for a rep to review, rather than sending anything unsupervised.
- Stage 6 (write the message): the same agent layer drafts messages grounded in that research, with the rep still owning voice and the send decision.
- Stage 7 (send): Sequencing runs email, calls, and social as one sequence instead of three disconnected tools.
- Stages 2 through 7 end to end: Plays can trigger the workflow from a signal match straight through to a sequenced send.
- Stages 8 and 9 (reply and handoff): these stay rep-owned. Unify surfaces reply classification and routing context, but the judgment calls, what to say back and which meeting matters, are left to the person, not the agent.
The honest limitation: stage 1, defining the target market, isn't something any tool can do for you. Unify can operationalize an ICP once it exists, but a wrong ICP just makes every agent-run stage downstream efficient at the wrong thing. The value case here isn't "more automation," it's fewer handoffs: continuous agent ownership of stages 3, 5, and 6 removes exactly the handoff points where teams currently lose the most time, while the stages that need judgment rather than throughput, 1, 8, and 9, stay with the rep.
That pattern shows up directly in CandorIQ's story. Founding SDR Zach Dettlinger was running outbound across four disconnected surfaces (Apollo for lists and sequencing, LinkedIn Sales Navigator for lookups, a separate web-intent tool, and Claude for drafting) before consolidating into one agentic engine. The result: $1.8M in pipeline attributed to Unify, 95% less time spent on manual tasks, an 87% lower bounce rate, and a 3.4% reply rate that was still climbing. "You're taking my time out of Claude, which is a beautiful thing," he said. "When I signed up, I would have never thought about that."
Sign up for Unify to see stages 2 through 7 of this map run from a single chat, with a rep still deciding what gets sent.
What Does This Look Like for One Account, Start to Finish?
Anrok, a sales tax compliance platform, is a useful worked example because their starting problem was the handoffs themselves, not any single stage. Before consolidating, sellers juggled three disconnected platforms (Outreach, LinkedIn Sales Navigator, and ZoomInfo) with no shared system for change management as the company entered new markets.
- Stage 1: Anrok's ICP centers on finance and tax teams at companies expanding into new state or international tax jurisdictions.
- Stages 2 to 4: account lists built and enriched against that ICP, with contacts verified through Unify's data layer instead of a separate tool.
- Stage 3, ongoing: accounts re-ranked continuously using signals including new hires, champion tracking, lookalike accounts, and website visitor behavior.
- Stage 5 and 6: AI personalization tied messaging to the specific trigger, a new hire in a tax-ops role, or a lookalike match to an existing closed-won customer, instead of a generic template.
- Stage 7: sequences run across marketing and SDR channels from one system instead of three.
- Stage 8: replies land in one inbox instead of split across Outreach and Sales Navigator.
- Stage 9: meetings route within the same system that ran the play, so context isn't rebuilt from scratch.
The outcome, per Anrok's published customer story: $300K+ in pipeline within the first three months, SDR workflows 4x faster than the prior ZoomInfo and Outreach setup, campaigns built 20% faster than in HubSpot, and three tools consolidated into one system. "Unify helped us build a complete outbound motion that actually drives revenue," said Kathleen Kong, Growth Marketing Lead at Anrok. "It's faster, smarter, and more connected."
What Changes at Different Team Shapes?
The nine stages don't disappear as a team gets smaller, they collapse onto fewer people. What changes is who owns each handoff, not whether the handoff exists.
Where Do PLG Signups, AE-Owned Motions, and Multi-Product Teams Fit This Map?
A few motions don't start at stage 1, and treating them like they do is a common source of wasted effort.
- PLG signups enter at stage 5, not stage 1. Someone already found the product themselves. The workflow starts at research (why did this account sign up, what are they doing with it) rather than at target-market definition or list building.
- AE-owned outbound collapses stages 2 through 4 into one person. Without a dedicated SDR, the AE personally builds, prioritizes, and verifies their own patch, which is why the list is usually smaller and worked more deeply than a team-wide motion. See AE-Owned Outbound Without SDRs.
- Multi-product companies run two workflows on the same account, not one. The same company can sit at stage 3 (prioritize now) for one product and stage 1 (not yet in market) for another. Treating both as a single motion causes cross-messaging collisions.
- Champion re-engagement re-enters at stage 4, not stage 1. A known contact who changed jobs skips target-market definition and list building entirely; verification and research are the actual work.
- Closed-lost re-engagement is a stage 3 decision, not a stage 6 one. The account already has message history. The real question is whether it re-enters the prioritized list, not what the new email says.
What Are the Signs a Stage Is Being Skipped Entirely?
These are the tells that a stage isn't just underperforming, it's not happening at all.
What Are the Most Common Mistakes Teams Make?
- Rewriting the message when the real failure is upstream in prioritization. A better opener doesn't fix a list built from the wrong accounts.
- Treating list size as a proxy for pipeline. Unprioritized volume just produces more low-relevance sends, not more meetings.
- Letting research and drafting happen as two disconnected steps. Different people, or the same person at different times, and the connection between them gets lost.
- Leaving replies in a shared inbox with no named owner. Positive responses go stale before anyone answers them.
- Routing meetings by rep availability instead of account ownership. This breaks continuity the prospect already built with a specific person.
Frequently Asked Questions
What does a good outbound workflow look like end to end?
A good outbound workflow runs through nine stages: define the target market, build the account list, prioritize the accounts, find and verify the people, research the account and person, write the message, send across channels, handle the reply, and hand off to a meeting. Each stage produces a specific artifact that gets passed to the next one, and most breakage happens in the handoff between two stages rather than inside a single stage.
How do I set up a cold outreach process?
Setting up a cold outreach process means running the first seven stages of the workflow in order at least once: define who you're targeting, build the account list, prioritize it by fit and signal, find and verify the people on it, research each account and contact, write the message, and send it across the right channels. Teams that can do this end to end in hours rather than weeks usually have the ICP and signal criteria decided before they touch a tool.
How do I go from an account list to booked meetings?
Going from an account list to booked meetings means running stages three through nine: prioritize the list by fit and live buying signal, find and verify the people on the prioritized accounts, research each one, write and send the message, handle whatever reply comes back, and hand off to a meeting with the account's context intact. The most common place this breaks is the handoff into stage nine, where a good reply never turns into a routed, contextualized meeting.
Which stage of the outbound workflow causes the most breakage?
It's rarely a single stage. The most common breakage points are the handoffs: research (stage 5) that never makes it into the message (stage 6), and a positive reply (stage 8) that never gets routed to a booked, contextualized meeting (stage 9). Teams tend to notice the stage they can see, usually the message, and rebuild that first even when the actual failure sits one stage upstream or downstream.
Do I need a dedicated SDR to run all nine stages?
No. The nine stages don't disappear without an SDR, they just collapse onto fewer people. A solo founder runs all nine personally. An AE running outbound without an SDR typically merges list building, prioritization, and verification into their own workflow and works a smaller, more deeply researched list. What changes by team shape is who owns each handoff, not whether the stage happens.
How is this different from a sales engagement platform?
A typical sales engagement platform owns stages six through eight: writing, sending, and tracking replies. It usually doesn't own list building, prioritization, or contact verification, which is exactly why teams end up stitching together a separate data tool, a separate enrichment tool, and a separate engagement tool, and why the handoffs between those tools become the weakest points in the whole motion.
Can AI run the entire outbound workflow without a human?
No, and the stages that most need a human are the first and the last two. Defining the target market is a judgment call no model should make unsupervised, and handling a reply plus deciding how a meeting gets routed both depend on context an agent doesn't have. Agents can reliably prioritize, research, and draft; a rep should still own deciding, sending, and replying.
How often should an account list be refreshed?
There's no single published cadence worth citing here, and refresh frequency by itself is the wrong metric to chase. The better test is whether the top of your prioritized list is actually changing week to week as new signals come in. A list that looks identical for three months straight means stage three has stalled, regardless of how recently it was technically rebuilt.
Glossary
- Outbound workflow: the full sequence of stages a team runs to turn a target market definition into a booked, qualified meeting.
- Play: an automated workflow that triggers a sequence of actions, research, enrichment, sequencing, from a specific signal or event.
- Account list: a set of companies matching a defined ICP, before any ranking by fit or intent has been applied.
- Prioritization: ranking an account list by a combination of firmographic fit and live buying signal, so reps work the highest-value accounts first.
- Enrichment: the process of attaching verified contact and firmographic data, email, phone, title, company details, to a raw account or contact record.
- Sequence: a defined series of outbound touches across one or more channels, sent to a contact over a set period of time.
- Handoff: the point where the output of one stage becomes the input to the next; most outbound breakage happens here, not inside a single stage.
- Reply classification: sorting an inbound response into a category, such as positive, objection, referral, or unsubscribe, so it can be routed to the right next action.
- Meeting handoff: routing a booked meeting to the correct owner with enough account and conversation context that the prospect doesn't have to repeat themselves.
Sources
- Unify customer story, Anrok: https://www.unifygtm.com/customers/anrok
- Unify customer story, CandorIQ: https://www.unifygtm.com/customers/candoriq
- Unify blog, "Introducing Lists and One-off Tasks for Human-in-the-Loop Outbound," Mar 25, 2026: https://www.unifygtm.com/blog/introducing-lists-and-one-off-tasks-for-human-in-the-loop-outbound
- Unify product page, B2B Company & Contact Data: https://www.unifygtm.com/product/b2b-company-contact-data
- Unify product page, Agents: https://www.unifygtm.com/product/agents
- Unify product page, Signals & Intent: https://www.unifygtm.com/products/signals
- Unify product page, Sequencing: https://www.unifygtm.com/product/sequencing
- Unify product page, Plays: https://www.unifygtm.com/product/plays
- Unify /explore, "What Is Outbound Sales? A Guide for GTM Teams": https://www.unifygtm.com/explore/what-is-outbound-sales-a-guide-for-gtm-teams
- Unify /explore, "Best B2B Prospecting Tools for Targeted List Building": https://www.unifygtm.com/explore/b2b-prospecting-tools-targeted-list-building
- Unify /explore, "Best AI Tools to Build and Prioritize a Target Account List": https://www.unifygtm.com/explore/best-ai-tools-build-prioritize-target-account-list
- Unify /explore, "8 Best Email Finder Tools for Verified Emails": https://www.unifygtm.com/explore/best-email-finder-tools-verified-emails
- Unify /explore, "How AI Agents Research Prospects": https://www.unifygtm.com/explore/how-ai-agents-research-prospects
- Unify /explore, "How to Personalize Outreach at Scale Without Sounding Like AI": https://www.unifygtm.com/explore/ai-outreach-without-sounding-like-ai
- Unify /explore, "LinkedIn and Phone Steps in Outbound Sequences": https://www.unifygtm.com/explore/multi-channel-outbound-linkedin-phone-context
- Unify /explore, "Automate Reply Classification & Follow-Up in Outbound": https://www.unifygtm.com/explore/automate-reply-classification-follow-up
- Unify /explore, "Meeting Routing for Signal-Led Outbound": https://www.unifygtm.com/explore/meeting-routing-signal-led-outbound
- Unify /explore, "From ICP to Live Outbound Sequence in Under a Week": https://www.unifygtm.com/explore/icp-to-live-outbound-sequence-six-hours
- Unify /explore, "How to Build Outbound Without an SDR Team": https://www.unifygtm.com/explore/outbound-without-sdr-team-playbook
- Unify /explore, "AE-Owned Outbound Without SDRs": https://www.unifygtm.com/explore/ae-owned-outbound-without-sdr
- Unify /explore, "How to Structure a Sales Team for Signal-Based Outbound": https://www.unifygtm.com/explore/signal-based-outbound-team-structure
- Unify /explore, "How Many Follow-Ups Should a Cold Email Sequence Include": https://www.unifygtm.com/explore/how-many-follow-ups-cold-email-sequence
- Unify /explore, "When to Retire an Outbound Sequence": https://www.unifygtm.com/explore/when-to-retire-an-outbound-sequence
- Unify /explore, "Automated Outbound Metrics: Three-Tier Framework": https://www.unifygtm.com/explore/automated-outbound-metrics-to-track
Austin Hughes is Co-Founder and CEO of Unify, outbound AI for sellers where AI agents and reps work side by side, from finding the buyers already in market to reaching them with the right message. 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.




