The phrase "let's run outreach on 20 accounts" is the most expensive sentence in B2B sales planning. It is almost always picked from intuition, not math, and it determines the rest of your stack for the next year. Capacity planning is not a forecasting exercise — it is a backwards calculation from the pipeline target, and once you have run it once, most of the operational decisions (how many accounts, what tools, what budget) become obvious. This guide is the calculation.
Most teams plan capacity in the wrong direction
The wrong direction: "We have 5 accounts available, each sends 80 messages a week, so we can do 1,600 messages a month." Then they hope the pipeline lands somewhere reasonable. It usually does not, and they blame the message.
The right direction: "We need 30 booked meetings a month. At our conversion rates, that requires X replies, Y connections accepted, Z requests sent, which at safe per-account volumes is N accounts." Now the capacity matches the target, not the other way around.
The right direction — work backwards from pipeline
Five steps, in order:
- Meetings needed per month (the input).
- Replies needed = meetings ÷ reply-to-meeting conversion.
- Connections accepted needed = replies ÷ accept-to-reply conversion.
- Requests sent needed = accepted ÷ acceptance rate.
- Accounts needed = requests ÷ safe per-account weekly volume × weeks.
Every step has a real conversion rate. Use realistic numbers, not optimistic ones.
Realistic numbers at each conversion step
| Step | Conservative | Average | Strong |
|---|---|---|---|
| Acceptance rate | 22% | 32% | 45% |
| Reply rate (of accepted) | 10% | 20% | 35% |
| Reply-to-meeting | 25% | 40% | 55% |
| Safe new connections / account / week | 50 | 80 | 100 |
Use Conservative for planning if you have no historical data; Average if you have 3+ months of clean numbers; Strong only if you have proven you operate at top-quartile across all three stages.
A worked example — 30 meetings/month
Target: 30 booked meetings per month from cold LinkedIn outreach. Using Average conversion rates:
Meetings target 30 / month ÷ reply-to-meeting (40%) = Replies needed 75 / month ÷ reply rate (20%) = Acceptances needed 375 / month ÷ acceptance rate (32%) = Requests needed ~1,172 / month ÷ ~4.3 weeks = Requests / week ~273 ÷ 80 per account safe = Accounts needed ~3.4 → round up to 4
So 30 cold-LinkedIn meetings a month is a 4-account motion at Average conversion rates. Drop conversions to Conservative across the board and the same target requires ~9 accounts.
The safety ceiling per account
The "safe weekly volume per account" line in the model is the most-abused number in this calculation. Operators with new or low-trust accounts who push 100+ connection requests a week run head-first into LinkedIn's rate limits — and once an account is restricted, capacity collapses, not increases.
Per-account safe ceilings, by account quality:
| Account state | Safe / week | Note |
|---|---|---|
| Fresh (< 3 months, no warmup) | 15–25 | Ramp slowly; high ban risk |
| Warmed (3–6 months, warmed properly) | 40–70 | Safe operating range |
| Aged (12+ months, real activity) | 80–120 | Standard for serious operators |
| NFC-verified, 24m+ aged | 100–150 | Highest sustained volume; lowest ban probability |
If your model says you need 9 accounts but your accounts are all fresh, you need either more accounts or better accounts. Aged/verified accounts can cut the count in half — see the cost-of-quality math in buy vs rent cost breakdown.
Buy your accounts — $350 once, yours forever.
NFC passport-verified, 2+ year aged, with 500+ targeted connections — owned, not rented. Up to ~71% cheaper than renting over a year.
See the buy offer →When to add accounts vs. when to fix conversion
The most common mistake at this stage: assuming the answer is always "more accounts". Sometimes it is — but often the higher-leverage move is to improve one of the conversion rates.
| Problem | Add accounts? | Or fix conversion? |
|---|---|---|
| Acceptance rate < 25% | No | Fix profile + note copy first |
| Reply rate < 8% on accepted | No | Fix targeting + first-message copy |
| Reply-to-meeting < 25% | No | Fix the booking/qualification step |
| All conversions healthy, output below target | Yes | Add accounts (the model says so) |
The point of the model is to expose which lever to pull. Adding accounts before fixing conversion is one of the most expensive mistakes in this space.
A spreadsheet template you can copy
Build a one-sheet model with the following columns:
- A: Meetings target (input)
- B: Reply-to-meeting conversion (input)
- C: Replies needed = A / B
- D: Reply rate (input)
- E: Acceptances needed = C / D
- F: Acceptance rate (input)
- G: Requests needed = E / F
- H: Requests per week = G / 4.3
- I: Safe per-account weekly volume (input)
- J: Accounts needed = ROUNDUP(H / I, 0)
Update inputs B, D, F monthly from your real numbers. The model becomes the operational source of truth for capacity decisions — see the KPI dashboard for how to feed it.