Cold calling KPIs: what to track and how to calculate each one
Track the chain from dial to deal: call attempts, connect rate, conversation rate, conversation-to-meeting rate, meetings per 100 dials, show rate and closing rate, plus the reasons behind every not interested. Split each number per caller and per campaign. Compare with your own four-to-six-week baseline, because external benchmarks vary widely and use different definitions.
Which cold calling KPIs should a sales manager track? Follow the chain from dial to deal: call attempts, connect rate, conversation rate, conversation-to-meeting rate, meetings per 100 dials, show rate and closing rate. Add the reasons behind every "not interested". Look at each number per caller and per campaign, and compare it with your own history rather than with a benchmark from another market.
Each KPI answers one question. Together they show where the funnel leaks, so you fix the right thing: the list, the timing, the script or the offer. Below you find the formulas, a worked example and what to do when a number is low.
The cold calling KPIs at a glance
First agree on definitions, or every report turns into a discussion. These work well for most teams:
- Attempt: every dial, answered or not.
- Connect: someone picks up. Some teams count only the prospect, others also the receptionist. Pick one and stick to it.
- Conversation: you speak with the right person and get past your opening line.
- Meeting: a booked appointment with a date, a time and a calendar invite.
| KPI | Formula | What it tells you |
|---|---|---|
| Call attempts | Number of dials | Effort and capacity |
| Connect rate | Connects ÷ attempts × 100 | List quality and timing |
| Conversation rate | Conversations ÷ connects × 100 | Right contact person, strength of your opener |
| Conversation-to-meeting rate | Meetings booked ÷ conversations × 100 | Message, objection handling, the ask |
| Meetings per 100 dials | Meetings booked ÷ attempts × 100 | Overall efficiency, the planning number |
| Show rate | Meetings held ÷ meetings booked × 100 | Meeting quality and confirmation |
| Closing rate | Deals won ÷ meetings held × 100 | Fit and the rest of your sales process |
| Not-interested reasons | Count per reason ÷ all not interested × 100 | What the market tells you |
How to calculate each KPI and what to do when it is low
Call attempts
Count every dial per caller, per day and per week. It is the only KPI a caller fully controls. As a reference, The Bridge Group's 2025 research among 351 B2B companies found that sales development reps on phone-centric teams average 56 dials and 4.6 quality conversations a day.
Low? Look at the calendar before you look at the person. Admin, internal meetings and list research often eat calling time. Protect fixed calling blocks and prepare the list the day before.
Connect rate
Connect rate = connects ÷ call attempts × 100
In Gong's analysis of more than 300 million cold calls, the average rep connected on 5.4% of dials and the top quarter on 13.3%. Gong counts only calls answered by the prospect, so a rate that includes receptionists will be higher.
Low? It is usually the data or the timing, not the caller. Track invalid numbers as a separate outcome, check how many switchboard numbers are on the list, and compare connect rates by hour of the day.
Conversation rate
Conversation rate = conversations ÷ connects × 100
This shows how often an answered call turns into a real conversation with the right person.
Low? You are stuck at the gatekeeper, the list has the wrong contact people, or the first ten seconds lose people. Listen to five calls that ended early. Then rework your opener with our cold calling script guide.
Conversation-to-meeting rate
Conversation-to-meeting rate = meetings booked ÷ conversations × 100
This is the core skill metric. Gong reports a set rate of 4.6% of conversations for the average rep and 16.7% for top reps. That gap shows how much skill and message matter once someone is on the line.
Low? Check three things. Does the caller actually ask for the meeting? Is the step small enough, such as a 20-minute call instead of a full demo? Are the answers to common objections ready? Our guide to handling sales objections helps with the last one.
Meetings per 100 dials
Meetings per 100 dials = meetings booked ÷ call attempts × 100
This combines the three rates above into one number. It is the best KPI for planning, because it tells you what one meeting costs in dials. At 1.5 meetings per 100 dials, one meeting takes about 67 dials.
Low? Look at connect, conversation and meeting rate to see which step drags it down. Asking for more dials rarely fixes it.
Show rate
Show rate = meetings held ÷ meetings booked × 100
Low? Send the calendar invite during or right after the call, add a short agenda and confirm the day before. If one caller has many no-shows, their meetings may be booked under pressure. Count rescheduled meetings separately from no-shows.
Meeting-to-deal (closing) rate
Closing rate = deals won ÷ meetings held × 100
Measure it per cohort: take the meetings held in March and follow how many became customers, even if they sign in May. Otherwise a long sales cycle makes the number jump from month to month.
Low? Check fit first. If meetings from one campaign rarely close, the target group or the promise made on the phone may be wrong. If every campaign is low, look at the meeting itself and the proposal.
Reasons for "not interested"
Give callers a short, fixed list and make them pick one reason for every "not interested":
- Already has a supplier
- No need at the moment
- No budget
- Not the decision maker
- Bad experience with similar offers
- No reason given
The share per reason is free market research. If "already has a supplier" is your top reason, you need a sharper comparison argument. A large "no reason given" group usually means callers do not ask why.
A worked example: from 1,000 dials to 2 customers
The numbers below are made up, to show how the KPIs connect. Replace them with your own.
| Step | Count | KPI |
|---|---|---|
| Call attempts | 1,000 | |
| Connects | 150 | Connect rate 15% |
| Conversations | 100 | Conversation rate 67% |
| Meetings booked | 10 | Conversation-to-meeting rate 10%, or 1 meeting per 100 dials |
| Meetings held | 8 | Show rate 80% |
| New customers | 2 | Closing rate 25% |
In this example, one new customer costs 500 dials. A team that wants one new customer per week needs about 500 dials a week, or 100 a day. With two callers, that is 50 dials each per day.
Now see what one better rate does. If the conversation-to-meeting rate rises from 10% to 12%, the same 100 conversations give 12 meetings, about 9.6 held and 2.4 customers. One customer then costs about 417 dials instead of 500. Coaching one step often beats adding more dials.
For the full calculation, read how many calls it takes to win one customer.
Track every KPI per caller and per campaign
A team average hides the useful information. Split every KPI two ways:
- Per campaign: if the same callers have a much lower connect rate on one campaign, the list is the problem, not the people.
- Per caller: if one caller has a far lower conversation-to-meeting rate on the same list, that is a coaching topic.
Be careful with small numbers. A rate based on 15 conversations can swing wildly from week to week. As a rule of thumb, wait for at least 30 conversations or a few hundred dials before you draw conclusions about a person or a campaign.
Use the numbers to start a conversation, not to judge. Our article on coaching a sales team with data shows how.
Why you should benchmark against your own history
Generic benchmarks are tempting, but they vary a lot:
- In Gong's data, the connect rate ranges from 5.4% for the average rep to 13.3% for the top quarter, within the same data set.
- Cognism's 2025 cold calling report puts the industry's cold call success rate at around 2%, against 6.7% for Cognism itself, without defining exactly what it counts.
- The same report finds that three call attempts cover 93% of conversations, while RAIN Group's research counts an average of 8 touches to get a meeting.
Definitions differ too. One source counts the receptionist as a connect, another only the prospect. One measures meetings per dial, another per conversation. And each data set comes from the vendor's own customers or team, often in other markets than yours. A Flemish installer or a Dutch accounting firm will see different numbers.
Your own history is the better benchmark:
- Run a new campaign for four to six weeks and record the KPIs. That is your baseline.
- Compare each week with the average of the previous four weeks.
- Compare campaigns and callers with each other, using the same definitions.
- Set targets as an improvement on your own baseline, for example two extra meetings per 100 conversations.
External figures still work as a sanity check. If your connect rate sits far below the ranges above, look at your data first.
A 15-minute weekly KPI review
- Call attempts per caller, against plan.
- Connect rate per campaign, against the four-week average. Down? Check invalid numbers first.
- Conversation-to-meeting rate per caller. Listen to two calls from the lowest score.
- Meetings per 100 dials: rising or falling?
- Show rate of last week's meetings, and the reason behind each no-show.
- Top three reasons for "not interested", and one change to the script or the arguments.
- One concrete action per caller for next week.
Collecting these numbers by hand from spreadsheets often takes longer than the review itself. Ringlane builds its reports from the team's own calls: call attempts, connect rate, conversations, conversation-to-meeting rate, meetings as a share of all contact attempts, a funnel and the reasons for "not interested", per member, per team and per campaign.
A rhythm calculator then shows how many calls per week and per day one new customer per week takes, based on your closing rate and your real connect and meeting rates.
Cold calling KPIs in short
Measure the whole chain, from attempt to customer, with fixed definitions. Split it per caller and per campaign. Fix the weakest step before you ask for more dials, and judge progress against your own baseline. The goal of a KPI is not a nicer report. It is knowing what to change on Monday.