How to Build a Repeatable Sales Model a New Rep Can Run

    Short answer

    The build order for a first transferable B2B sales model, from list and replies to a new-rep proof test and a gate before you scale.

    Artem Smirnov
    Artem Smirnov

    Last updated · 13 min read

    Artem Smirnov in a dark suit against a charcoal studio backdrop, next to the line 'WRITTEN DOWN IS NOT REPEATABLE. Someone else has to run it'

    A repeatable sales model is a written set of stages, from the first list you contact to the signed contract. Every stage has an entry rule, an exit rule, one metric and one named owner. The model is repeatable when someone other than the founder runs it and gets similar conversion at each stage.

    You build it in the order a buyer meets you: buyer and offer, list, messages, replies, booking, handover and the close. Volume goes on last. Always last.

    My team's job is filling the calendar, which in plain terms is appointment setting: we reach buyers by cold email and LinkedIn messages, and we do no cold dialing at all. The client's own people take those meetings and close the deals. I write about the top half of the model from inside it, and on closing I stick to what the stage numbers show.

    One month shows why the stages matter more than the totals. One LinkedIn account over 30 days reached 1,200 people. Of them, 289 replied, and 21 of those were interested. Then 16 booked a meeting, 12 showed up and 3 closed. I use this month as my reference funnel because every stage shows up in it.

    What is a repeatable sales model, and how do you know you have one?

    Each of the four parts does one job:

    • The entry rule says when a contact or deal arrives in a stage.
    • The exit rule says what has to happen before it moves on.
    • The metric is the one ratio that shows whether the stage works.
    • The owner is the person who answers for that ratio.

    A written process is not the test. Look at who runs it. If only the founder gets the numbers, you have a habit, not a model. Much of what works lives in the founder's name, network and judgment on the call. You have proof on the day a second person runs the model and the ratios hold.

    Two terms get mixed up here. A sales process is the stages and the rules between them: what happens, in what order, and who owns it. A sales methodology is how a person behaves inside one stage, such as the way a closer questions a buyer.

    The stages before the call belong in the model because many buyers do their checking before they talk to anyone. In a Gartner survey run in August and September 2025, 67% of 646 B2B buyers said they prefer a "rep-free experience" (Gartner).

    From the outbound side, I put it bluntly: nobody buys from an email. They buy after they audit you, long before any call. Your list, your message and your reply handling (the top half of B2B lead generation) shape that check. That makes them stages with owners, the same as the proposal.

    A common founder mistake is to hire three salespeople and hope a model appears. Mark Leslie and Charles Holloway warned against this in Harvard Business Review in 2006. In their view, a company that ramps sales capacity right away burns cash and misses its revenue targets. The whole organization first has to learn how customers buy and use the product (HBR).

    In what order should you build the model?

    Here is the full sequence, with what each step has to produce:

    1. Buyer and offer. One audience and one clear thing to say yes to.
    2. List. The right companies, and the titles that actually decide.
    3. First message and follow-ups. A full sequence with several touches.
    4. Reply handling. Someone reads, tags and answers every reply.
    5. Booking. Interested replies become dated meetings.
    6. Handover. The closer gets what they need before the call.
    7. Call to close. Proposal, decision, signature.
    8. Volume. More contacts, more accounts, more inboxes.

    The order follows a rule my team works by. Fix the funnel first. Then scale. Volume is the last step. If it does not convert small, it will not convert big. So the first thing we check is the stage ratios on a small first batch. If the model leaks at reply handling, more contacts only produce more ignored replies.

    One LinkedIn account produced the month above. That is enough to test steps 1 to 7 before you pay for step 8.

    Founder-ledDocumentedScaled
    Who runs itThe founder runs every stageA second person runs at least one stage while the founder watches the numbersA named owner runs each stage; the founder reviews ratios
    What is written downLittle or nothingOne page per stage with entry and exit rulesStage pages, handover template, message library, onboarding plan
    Main metricDeals closed this monthEach stage ratio, compared with the founder'sStage ratios per owner, plus win rate and cycle length
    Main riskNobody else can repeat itExit rules that describe feelings instead of eventsAdding reps or inboxes before the ratios hold

    What should each stage's entry and exit criteria look like?

    Buyer and offer are decisions you make before anyone enters a stage, so the table starts at the list. Every row gets exactly one owner.

    StageEntry ruleExit ruleMetricOwner
    ListThe company fits the buyer you described in one sentenceRight title found, address verifiedBounce rateList builder
    SentVerified contact loaded into the sequenceContact replied, or the sequence endedReply rateCampaign owner
    RepliedAny reply arrivesReply tagged (interested, not now, wrong person, no) and answered within one working dayReplied to interestedInbox manager
    InterestedReply asks about price, timing or fit, or says yes to a callA meeting time is proposedInterested to bookedBooker
    BookedCalendar invite acceptedHandover note delivered to the closerBooked to showedBooker
    ShowedBuyer joins the callNext step dated in the calendar, or a recorded noShowed to proposalCloser
    ProposalBuyer asks for a price or a proposalSigned, or lost with a written reasonProposal to closedCloser
    ClosedContract signedHanded to delivery with the sales notesWin rate, cycle length in days from first reply to signatureCloser

    Write every exit rule as an event someone can check. "Replied with a question about price" passes. "Seems keen" fails, because two people will read the same reply two ways.

    Copy this template into a doc or a CRM field description:

    A contact leaves [stage] when [observable event]. [Owner] records it in [field] within [time]. If it has not happened after [time], the contact moves to [fallback].

    Here it is filled in for the booked stage: a contact leaves "booked" when the closer has the handover note. The booker records it in the deal notes the same working day. If the note is still missing the day before the call, the meeting moves to the founder's review list.

    Two owners on one stage means each can assume the other did the work. The person who books a meeting and the person who wins the contract hold separate jobs. Put different names on the booker and closer rows, even if the founder still sits in both seats today.

    Where do replies and handover fit in a sales process?

    Between "sent" and "booked" sits the stage most models skip. A reply that sits for two days goes cold, and that reply was the one thing the whole campaign was paid to produce.

    My team puts a named inbox manager on replies so none slips through. That person reads each answer, tags it and responds the way a human would. The next move is covered in how to move from a positive reply to a booked call.

    Set expectations from real ranges. These are what my team sees across our own email campaigns:

    • A reply rate of about 1% is usual. Once you send thousands of emails a day, expect closer to 0.5%.
    • Usually 10% to 50% of replies turn into meetings.
    • A narrow list with deeply researched messages can climb to 15% to 17% replies, and about 75% of those people turn out interested.

    The averages fall as the volume climbs. Treat these ranges as a sense of scale. Do not copy them into your targets or read them as a promise for your company.

    The sender is part of the model too. The same email, word for word, can land a very different reply rate depending on who sent it. Buyers look up whoever wrote to them before they answer. When a new rep takes over a stage, their LinkedIn profile becomes part of what they own.

    The handover is a stage with its own form. Before every call, the closer gets a note with:

    1. Who: name, title, company, and who else is likely to weigh in
    2. How they arrived: list segment, and which message or follow-up they answered
    3. What they said: the reply, copied word for word
    4. Why they booked: the problem or question they raised
    5. What they have seen: any case study, video or page sent to them
    6. Open points: price, timing or fit questions the closer must cover
    7. What they expect from the call, in their words

    The person taking the calls needs real room in their calendar. If they keep postponing meetings, the system stalls, however good the ratios are upstream.

    What does a working funnel look like in real numbers?

    Take the month from the top of the page and divide each stage by the one before it:

    StepRatioOwnerA drop here usually points to
    Contacted -> repliedabout 24%Campaign ownerWrong list, weak sender or weak first message
    Replied -> interestedabout 7%Inbox managerSlow answers, or an offer that fits too few people
    Interested -> bookedabout 76%BookerSlow scheduling, too much back and forth
    Booked -> showed75%BookerWeak confirmation, a long gap before the call, little to find when they check you
    Showed -> closed25%CloserCall structure, proposal or fit

    From contacted to closed, that is one deal for every 400 people reached. A founder looks at the 3 signed contracts. I look at the 9 people who showed up and did not sign.

    Do not hold that 24% against the email ranges above. It came from one LinkedIn account at small volume, and averages fall as volume grows.

    The biggest drop is replied to interested: fewer than 1 in 13 replies asked for a conversation. That ratio belongs to whoever writes the message and answers the inbox. This is why reply handling gets its own owner and its own number.

    At the closing end, one showed meeting in four became a deal. Booked meetings were 16 of 289 replies, about 6%. That gap traces back to the weak replied-to-interested step, not to the booker, who turned 16 of 21 interested people into meetings.

    A full calendar only pays when somebody closes, which is why booking and closing get separate owners in the table above. A model that counts calendar invites as wins hides the leak.

    If your closing ratio is low today, expect it to be lower still once more calls arrive. Piling on extra calls will not mask a weak close. It only exposes the weakness faster. Fix the closer's stage before you feed it more meetings.

    How do you test that a new person can run it?

    Run the proof test in this order:

    1. Write one page per stage: entry rule, exit rule, metric, owner, and the messages or notes that stage uses.
    2. Record the founder's ratio for every stage over a comparable period, such as the last full month.
    3. Give the new person a defined slice for a fixed window of at least a month. The slice can be one segment of the list or one stage.
    4. Keep the founder out. No rewriting replies and no joining calls to rescue them. Answer questions by improving the stage page.
    5. At the end of the window, compare the new person's ratios with the founder's, stage by stage.
    6. Log every question the new person asked. Each one marks a gap in the pages.

    The decision rule follows. If the new person's ratios stay within the founder's range at every stage, the model passes. If one stage drops, fix that stage only and rerun it. If several stages drop at once, the model still lives in the founder's head, and you go back to the pages.

    Agree on the range before the test starts. A workable tolerance is within a quarter of the founder's ratio. Treat that as a rule of thumb, nothing more. On a 75% show rate, for example, the new person passes at about 56% or above.

    Watch the sample size at the bottom of the funnel. In the month above, one extra deal would have moved the closing ratio by about 8 points. Judge the closer's stage over more calls than a single month may give you. For a reference point, here are meeting-to-close figures from real campaign windows.

    Keep the test simple enough to be fair. In a Gartner survey of 1,026 B2B sellers in 2024, 72% said they felt overwhelmed by the number of skills their job requires. Overwhelmed sellers were 45% less likely to hit quota (Gartner). Few stages, one tool and one page per stage give the new person a real chance.

    What does a 30-60-90 ramp look like for a new rep on the model?

    DaysWhat the rep doesWhat you measurePass mark
    1 to 30Shadows every stage, reads the pages, tags replies while the inbox manager checks themQuestions logged; reply tags matched against the inbox manager'sExplains every exit rule without notes
    31 to 60Owns one stage alone, usually replies or bookingThat stage's ratio against the founder'sRatio inside the agreed range
    61 to 90Runs the full model for their sliceEvery stage ratioThe proof test passes

    A ramp needs a steady flow of meetings to learn on, so check the calendar before day 1.

    Ninety days is a working assumption for a model with short sales cycles. Sequoia, writing on the sales learning curve in 2016, puts the typical enterprise rep ramp at 180 days (Sequoia). That is an enterprise example, so use it for enterprise plans only.

    When is the model ready to scale?

    Sequoia's sales learning curve has three staffing phases: initiation, transitional and execution. It starts with three or four reps. Its traction point is a rep producing 2X their fully loaded cost, which is an example figure from enterprise selling.

    Stanford GSB, writing on the same idea in 2006, puts the hiring rule plainly. Start with a small group of salespeople, and hire faster once they make a positive marginal contribution (Stanford GSB).

    Add volume or headcount only when all three of these hold:

    1. The proof test passed with someone other than the founder.
    2. Stage ratios held for a period you chose in advance, such as two full months.
    3. The closer has calendar room for more calls this month.

    If one fails, hold off. More volume will only make that failure louder.

    Once the gate is open, I scale sideways: more LinkedIn accounts or more email inboxes on the setup that already works. I never push volume through a new inbox before its warm-up is done.

    As planning math, 30 to 35 booked meetings in a month need a monthly list of 10,000 contacts or more. Your own ratios from the proof test decide the real number.

    Review every stage ratio once a month. That monthly view is how you see a model start to hold. One US consulting and coaching client is a good example. My team booked 27 meetings in month one, 35 in month two and 64 in month three, and 204 across four months.

    Keep the pages alive: rewrite a stage page whenever its ratio moves outside the range. A page nobody has touched in a quarter usually no longer matches what the team actually does.

    If you are still the only seller, start by running the top of the funnel yourself. If a process already exists and still leaks, sales process improvement goes deeper on fixing it stage by stage. If the model works and you want more revenue from it, ways to increase B2B sales ranks the levers by how fast they pay back.

    This week, pull last month's numbers and write down the five stage ratios from contacted to closed. The stage with the worst ratio is the first page you write.

    If your sales only work when you are in the room, book a call and describe your current stages. I will help you decide what to write down first.

    Questions founders ask about repeatable sales

    Does the model depend on one of the top sales methodologies?

    No. Challenger, SPIN or MEDDIC shape how a closer behaves inside the call, and the stages work with any of them. Let the closer keep the one they already use. Never switch methodology during the proof test, though, or you cannot tell whether the stage or the new method moved the closing ratio.

    How does a sales playbook differ from a sales process?

    The process is the skeleton: stages, rules, owners and ratios. The playbook is what people say and send inside each stage, such as message versions, reply templates, objection notes and the handover form. Write the process first. A playbook without exit rules gives a new rep good lines but no way to tell when a deal has actually moved.

    Can a solo founder with no team build a repeatable sales model?

    Yes, up to the proof test. A solo founder can write the stage pages, track the ratios and run the funnel at small numbers. The test itself needs a second person, even a part-time one. A freelancer who handles replies for one month is enough to show whether a single stage survives without you.

    Should the model live in a CRM or in a document?

    Use both, for different jobs. The CRM holds the stages, so every contact sits in exactly one of them and every ratio can be counted. The document holds the rules and the reasons, and a new rep reads it before touching anything. Which CRM you pick matters little, as long as nothing falls between two stages.

    Want to see what we would build for your company?

    I am Artem Smirnov. I help established B2B companies build and run the go-to-market system behind their pipeline, from outbound to AI search visibility. Book a call and we will see what fits yours.

    Artem Smirnov
    Artem Smirnov

    I help established B2B companies build and run their go-to-market systems. 9 years in outbound, 10,000+ campaigns.