Sequence capacity planning: the two numbers that decide your cadence

Sizing email and LinkedIn cadences against real sending capacity since 2024.

What is sequence capacity planning?

Sequence capacity planning is the arithmetic that establishes how many prospects your channels can physically carry inside your campaign window, done before any touch is designed. It takes your prospect count and your daily capacity per channel, and returns which channel may make first contact and how many new prospects a day you can enrol without the follow ups exceeding what you can send.

A cadence is a scheduling problem with two hard constraints, and each is a single division. Most teams design the touches first and meet the constraints in week three, when the sending tool starts quietly delaying steps.

How do you work out which channel can open? The Coverage Window

The coverage window is how many working days it takes to make first contact with every prospect on your list using one channel, and it decides whether that channel may open the sequence.

coverage window (working days) = prospect count ÷ daily ceiling on that channel

The design rule we run, and you may disagree with it: a channel may only open the sequence if it can make first contact with the whole list inside the first half of the window. A 30 calendar day window holds roughly 22 working days, so an eight day email coverage window fits. The half is ours, chosen so the later touches have somewhere to land. It is a planning convention, not a measured optimum.

One warning about the send rate in that example, and it applies to every figure below it. Twenty five sends per mailbox per day is a placeholder used to make the arithmetic legible. It is not a house default, it is not a measurement, and it should not be quoted as either. Substitute whatever your own sending configuration actually runs at, and every number downstream moves with it.

How many new prospects a day can a sequence actually absorb? The Intake Ceiling

The intake ceiling is the number of new prospects a day your sequence can enrol without the accumulating follow ups exceeding your send capacity.

intake ceiling = daily send capacity ÷ number of email touches in the sequence

Every worked figure below runs on that placeholder of 25 sends per mailbox per day, flagged above and repeated here because the numbers travel. Ten mailboxes at that placeholder rate is 250 sends a day, so a sequence with three email touches has an intake ceiling of 83 prospects a day rather than 250. Substitute your own send rate and every figure moves with it. What does not move is the shape: every prospect who entered on Monday is still consuming a send slot on Thursday and again the following week.

Teams set intake equal to capacity because capacity is the number the dashboard shows them, and it holds for about a week. Then required volume becomes a multiple of what the infrastructure can place, and the tool either silently delays touches or the sender pushes past its throttle. There is no alert for "your cadence is now four days behind its own design", which is why this failure is invisible.

Why does the daily send load peak in week two?

The daily load is the total number of actions your infrastructure has to place on a given date, summed across every cohort in flight, and it peaks when the first cohort's last touch lands on the same date as the newest cohort's first touch.

Same 2,000 prospect example, three email touches, entering at the ceiling of 83 a day, on relative days 1, 4 and 9. The shape of this table is the point. The figures in it are arithmetic on an assumed send rate, not a measurement, and they move the moment you substitute your own.

Day nine is the peak and also the steady state, and nothing after it gets worse, which is what a correctly sized plan looks like. Now set intake to capacity, 250 a day on the same assumption: days one to three look perfect, and by day nine demand is 750 sends against 250 of capacity. The first week of every over intaked campaign looks healthy, which is the trap.

What I got wrong about this, and what it cost

Skipping this arithmetic cost us campaigns we killed for the wrong reason. We run 28 standing agent desks across the fleet, and one produces the daily segment snapshot that feeds the kill decision, at the 0.5% positive on sends line. The full threshold ladder and the reasoning behind it belong to our sibling property allboundpros.io, which has not published yet.

For a stretch in the first year we killed campaigns the snapshot said had failed, and some had not. On one, the send logs showed the third touch landing days late for the back half of the list, because intake had been set at sending capacity rather than at the intake ceiling. The sequence in the report was not the sequence that ran, so we graded a scheduling failure as a copy failure.

Nothing in the stack flags this, because a sending tool reports what it sent, not what it was supposed to send and when. So the check moved to the front: capacity arithmetic before anyone writes copy, and the intake number in the brief next to the list size.

How do you sequence thousands of prospects with one LinkedIn seat?

You do not touch all of them on LinkedIn. You spend the seat on the slice that showed a signal, which is what makes one seat viable against a far larger list.

On a list where 5% of prospects show any engagement signal, the seat only serves that 5%, so it can support a list roughly twenty times the size its coverage window suggests. That is the arithmetic behind the WideNET motion: email opens across the full market, LinkedIn escalates onto prospects who did something.

When even the slice exceeds seat capacity, rank it: replied without a decision, then clicked, then opened more than twice. Where open tracking looks unreliable, drop the last two tiers, because escalating on a noisy signal spends your scarcest channel on people who never saw the email.

What do you do when the arithmetic says no?

Four levers, each with a formula and a real cost.

If the required mailbox count is several times what you have, that is a scope problem, not a cadence problem.

Which lever should you pull first?

Extend the window, unless there is a dated reason you cannot, then cut the list. That order follows from what each lever costs.

Shortening the sequence is the only lever that costs something you cannot get back, so do it last, and only if the angle was already weak.

The contrarian consequence, and the argument this whole site is built on: the published fourteen day sequence template is unrunnable at any real list size. Fourteen calendar days is about ten working days. Every prospect has to receive touch one inside roughly the first five for the later touches to have anywhere to land, and on the worked example above that means intaking 200 a day against a ceiling of 83. Change the send rate and the two numbers move together, so the conclusion holds across a wide range of assumptions. Every published template looks the same as every other one because none of them was ever divided by a sending ceiling. That is not a small criticism of the genre. It is the reason cadence content in this category contradicts itself: the templates were drawn for one prospect and published as though they were plans.

What this arithmetic cannot tell you

Capacity planning tells you what fits in a window and nothing else.

It cannot tell you whether the offer is any good, whether the list is qualified, or what reply rate to expect. Cadence is the third most important thing about a campaign, after targeting and offer, and a perfectly sized calendar carrying a message nobody wants produces perfectly scheduled silence. If three genuinely different angles have produced close to nothing across a few thousand sends, the calendar is not your problem.

Two modelling limits are ours. The load arithmetic assumes zero attrition, while real sequences shed prospects to replies, bounces and suppressions, so it overstates load, which is the safe direction for a capacity plan and still an overstatement. Public holidays are not modelled.

Who should skip this: with a few hundred named accounts the arithmetic never binds. It earns its keep above roughly a thousand prospects, or the moment a second channel joins the plan.

Put real dates on it

The Sequence Cadence Builder returns the coverage window, the intake ceiling, the day by day load calendar with the peak marked, and the four levers with your numbers filled in. Free, no signup, and it works with whatever sending tool you already pay for.

MultichannelPros is part of the Outbound Pros group, operated by Jānis Plūme.