Touch count is a control setting, not a magic threshold. Set it against the event you want, give each touch a place on the calendar, and give it a reason to exist. One benchmark puts the outbound sweet spot at three attempts, with up to six attempts in the cadence.1 Another reports an average of eight touches to get an initial meeting or another conversion with a new prospect.2 The figures answer different planning questions. Use the count as a test boundary, then judge it by the outcome and the touches people actually received.
Set the outcome
Decide what the cadence must produce before setting the count. That keeps the count attached to a business event instead of making it an activity target.
The reported eight-touch figure concerns getting an initial meeting or another conversion, not making the sale.3 Decide whether you are measuring replies, meetings, qualified conversations, or a later commercial outcome. Keep that definition in the cadence record so every review uses the same one.
A practical setup has one primary outcome and one stopping signal. A reply may stop the sequence while you handle the conversation, and a meeting may move the contact into a different process. If there is no response, let the cadence run until it reaches the review point or maximum you set.
Choose the touch count
Pick a starting range and decide when you will inspect performance. Give the sequence enough room to work while keeping the first version easy to learn from.
For an outbound cadence, use three attempts as the first benchmark and six as the upper planning limit.1 Treat three attempts as a review point, not an automatic stopping rule. If the outcome has not appeared, inspect the timing, channel mix, and message progression before extending the sequence.
Planned touches and realized touches can differ. One example used an eight-touch cadence, then modeled six touches because some contacts would be reached before the cadence ended.4 Compare conversion rates using both figures. A contact who replies early has not received the full planned sequence, so report the touches delivered as well as the touches designed.
One framework suggests fifteen sales touches over three weeks.5 Use that denser pattern as a deliberate test when a market needs more repeated exposure, and give every added touch a clear reason. A longer sequence should earn its length through additional opportunities to start a conversation.
Space the touches
Spacing turns a list of attempts into a cadence. Write the dates before launch so each touch has a planned place and you can review the sequence later.
Cadence is the timed sequence of touches and the gaps between them.6 Define those gaps explicitly. A schedule such as Day 1, Day 3, and Day 6 gives the sequence predictable structure and coverage.7 Setting spacing by feel without a consistent schedule creates a different experience from one contact to the next and makes results harder to interpret.8
Choose the channel as part of the spacing decision. One sequence starts with phone, email, and a LinkedIn touch.9 When you call, leave a voicemail that previews the email and states its subject line, then send the email.10 After two business days, follow up in the same thread with, "Any thoughts?"11
Use the calendar to separate touches that need attention from touches that repeat the same interruption. A phone attempt followed by a related email gives the person a way to engage through the channel that suits the moment. The next touch should add a reason to respond, answer a likely question, or make the decision easier.
Set the duration
Duration tells you how long the planned sequence occupies the calendar. Set it from the first activity to the last, then check whether the window gives the touches enough room to arrive at useful moments.
Duration is the time between the first and last activities in a cadence.12 An example cadence uses eight touches over the course of a month.13 That is a design window, not proof that every contact will receive every touch.
The planned window should reflect a deliberate view of buyer time.14 Ask when the person is likely to notice a new message and how much room should exist between attempts. A short window can compress several touches into noise. A long window can make the sequence lose continuity. Set the interval pattern first, then see what total duration it creates.
Record both the scheduled end date and the actual end date for each contact. Early replies, meetings, or other outcomes will shorten the realized run. Those early exits are part of the result, not a flaw in the cadence.
Review the cadence
Review the sequence when it should have produced enough signal to make a decision. Look at movement through the cadence, then change one design choice at a time so you can see what improved.
One reported pattern says that 80% of meetings happen after touch 5.15 Use that finding to inspect what happens after the first review point before cutting the sequence short. It does not cancel the outbound benchmark of three attempts. It shows that a meeting may arrive later than the first useful checkpoint.
For every contact, record the touch reached, the channel used, the response, and the outcome. Compare the planned count with the realized count. If people engage early, study the touches that came before the response. If the cadence runs to its end, inspect whether the later messages gave the person a new reason to continue.
Each touch should move the conversation forward instead of repeating the ask.16 Aim for momentum and keep noise down.17 The sequence should respect how buyers use inboxes, social feeds, and the phone.18
What not to do
These mistakes distort the count and make spacing harder to learn from. Keep them out of the first version you run.
- Do not set the gaps by feel without a consistent schedule.8
- Do not use a touch count for getting a meeting as though it were a count for making the sale.3
- Do not abandon a prospect after two or three attempts solely because those attempts produced no response.19
- Do not assume a good message can overcome timing that misses the moment, because the message then has no chance to work.20
Before launch, write the outcome, the review point, the maximum sequence, each date, and the reason for every touch. After the run, compare planned and realized touches against the outcome, then adjust the range or spacing that failed.