Formula to estimate how many conversations can be reopened per day without affecting the incoming queue

What this formula solves

Reopening old conversations recovers sales that were almost there: the contact exists, the conversation happened, and the reason it went cold is usually an oversight, not a rejection. The problem is that every reopening consumes minutes from the same team that handles what comes in today. If the reopening quota is defined by enthusiasm and not by capacity, the incoming queue starts to fall behind, first response times go up, and the business loses more on the other side than it recovered.

This page does not say who to reopen: that is already covered in the decision matrix of which conversations to reopen the next day and which to close. Here the problem is different and more limited: how many. The formula takes the team's real capacity in minutes, subtracts what the incoming queue already consumes, and distributes the rest among reopenings, with a buffer for unforeseen events.

The four parameters that go into the formula

Before calculating, you need to gather four data points. None of them requires a special tool: they come from the business's message inbox and a week of observation.

• Daily team capacity (C): real available service minutes per day, adding up everyone who responds. It is not the full workday: it is the effective time in front of the inbox, excluding meetings, administrative tasks, and breaks.
• Incoming load (E): minutes per day consumed by handling what comes in without reopening anything. It is estimated by multiplying the number of incoming conversations by the average resolution time.
• Time per reopening (R): average minutes required to work an old conversation, from when it is opened until it is closed or the next step is scheduled.
• Operational buffer (K): percentage of capacity reserved for unforeseen events, peaks, and conversations that run long. It is expressed as a decimal (0.15 = 15%).

The most common mistake is calculating C with the theoretical workday. If two people each work eight hours, the capacity is not 960 minutes: it is the portion of those hours that they are actually looking at the inbox. Measuring for a week and averaging gives an uncomfortable but usable number.

The formula for the daily reopening quota

The daily reopening quota (Q) is the available capacity after the incoming queue and the buffer, divided by the time of each reopening:

Q = [ C × (1 − K) − E ] ÷ R

It reads like this: from the total capacity, the buffer for unforeseen events is deducted first, then what the incoming queue already consumes is subtracted, and what remains is divided by the time of one reopening. The result is the number of old conversations that can be worked per day without touching the response time of what comes in.

If the result is zero or negative, the answer is not to adjust the formula: it means there is no capacity to reopen today. In that case there are three paths, and it is best to choose one consciously instead of forcing the number: expand the effective capacity, lower the time per reopening with a shorter opening message, or accept that reopening is done only in dead gaps and without a fixed quota.

Example with numbers to see how it moves

Let's suppose a team of two people who handle the inbox four effective hours each: C = 480 minutes. The incoming queue consumes E = 300 minutes per day. The buffer is K = 0.15. Each reopening takes R = 6 minutes.

Q = [480 × (1 − 0.15) − 300] ÷ 6 = [408 − 300] ÷ 6 = 108 ÷ 6 = 18. The quota is 18 reopenings per day. If the time per reopening rises to 10 minutes because the opening message invites a long conversation, the quota drops to 10.8, that is, 10 reopenings. The formula shows something that intuition hides: the incoming queue still takes up the same place, so the quota does not drop in the same proportion that the time per reopening rises, but more: it is almost halved when the time almost doubles.

The table reveals two things. First, the quota is highly sensitive to the incoming queue: between the first and second scenarios, with the same capacity and the same time per reopen, the quota drops from 20 to 2 only because the incoming queue grew. Second, the buffer is not decoration: in the large team with high queue scenario, raising K from 0.15 to 0.20 is the difference between a quota of 9 and one of 4. The buffer protects precisely the days when the incoming queue overflows.

How to measure each parameter without any fancy tools

1. 1Measure C over five business days: note in a spreadsheet each time block in which someone was actually responding to messages. Add up the blocks for the whole team and average. Do not count time when the inbox was open in the background.
2. 2Measure E with the same method: count the incoming conversations each day, take a sample of twenty and time how long each one takes from the first response to the close. The average of the sample multiplied by the daily total gives E.
3. 3Measure R on the reopens you have already done, even if there were few. If you have never reopened, use the time of a similar incoming conversation and add one minute for searching the history.
4. 4Define K according to your business variability. A business with predictable peaks (for example, Monday mornings) can use 0.10; one with irregular demand, 0.20 or more.
5. 5Calculate Q and always round down. A quota of 10.8 is 10 reopens, not 11.
6. 6Repeat the measurement every month. C, E and R change with the season, with vacations and with team changes.

What to do when the calculated quota is not enough

It is normal for the first calculation to give a small quota. The temptation is to ignore it and reopen anyway, but that shifts the problem to the incoming queue, where the cost is higher: a new customer waiting for a response is more expensive to lose than an old one who has already cooled off. There are four levers to raise the quota without breaking anything.

• Lower R: a short opening message, with a single question, reduces the time per reopening. Openings that explain too much invite long replies.
• Raise C in the right time slot: if the incoming queue has a clear peak, adding capacity outside that peak allows you to reopen without competing for the same minutes.
• Lower K only if you have data: reducing the buffer is valid when the measured variability is low, not when intuition says the business is stable.
• Reopen on alternate days: instead of forcing a daily quota, concentrate reopenings on the days with the smallest incoming queue. The weekly quota stays the same and the impact on the daily queue drops.

The fourth lever is usually overlooked. The formula calculates a daily quota, but nothing forces you to use it every day. If on Monday the incoming queue eats up the capacity, Monday's quota is zero and the reopenings move to Tuesday. What you cannot do is accumulate quotas: the quota of a day not worked is not transferred, because the next day's capacity is already committed.

Checklist to verify that the incoming queue was not affected

The quota is a hypothesis until it is tested. These are the signals to watch during the first two weeks of reopenings, in order of importance.

• First response time of the incoming queue: if it rises more than usual on days with reopenings, the quota is too high.
• Incoming conversations unanswered at the end of the day: if even one appears that was not there before, the quota went too far.
• Reopenings left halfway: if many are opened and not closed, R was underestimated and the real quota is lower.
• Distribution by operator: if a single person does all the reopenings, the team quota does not apply to that person and you have to calculate one for them.
• Reopenings that generate a response: if the response rate is very low, the problem is not the quota but the selection of whom to reopen.
• Changes in incoming volume: an active campaign or a high season invalidates the quota calculated the previous week.

Common mistakes when applying the formula

These are the four mistakes that make the calculated quota useless in practice.

• Using theoretical shift time as capacity: C is effective time at the tray, not contracted hours.
• Calculating E with the total number of messages instead of conversations: a conversation can have twenty messages and take ten minutes, or three messages and take two.
• Not separating reopening time from follow-up time: if a reopening turns into a long conversation, that time comes out of the next day's incoming queue, not out of the day's quota.
• Setting the quota once and never reviewing it: parameters change, and the quota that worked in March can be dangerous in December.

How it connects to the rest of the support system

The reopening quota is one piece within a larger calculation. The capacity used here is the same one that is distributed among the incoming queue, follow-ups and administrative tasks. If the business has already calculated how many conversations an agent can sustain per shift, that number is the ceiling for C. If it has also measured the backlog and its drain time, it knows how many days of reopenings lie ahead and can plan the quota based on that horizon.

In practice, most teams don't need a new tool for this: they need two weeks of honest measurement and a spreadsheet with the four columns. A message tray with AI suggestions, like Wando's, helps lower R because the opening reply is approved with a click instead of being written from scratch, but that doesn't change the logic of the calculation: it changes one of the parameters.

It's a good idea for the quota to be written down and visible to the team. A team that knows it can reopen ten conversations per day chooses them better, works them completely and doesn't neglect what comes in. A team that reopens whenever it can ends up doing both things halfway, and in customer support that costs you twice.