One reply from 20 messages fits a true reply rate anywhere from about 1 to about 24 percent, five from 100 fit about 2 to 11 percent, and 20 from 400 fit about 3 to 8 percent, using the Wilson interval that the NIST/SEMATECH e-Handbook of Statistical Methods gives for a proportion.
What a small sample can say
The table applies the Wilson interval at the 95 percent level to five sample sizes and five observed reply rates. It treats each message as an independent try at the same list, which holds only loosely when one person writes every message.
| Messages sent | 0 percent | 2 percent | 5 percent | 10 percent | 20 percent |
|---|---|---|---|---|---|
| 20 | 0 to 16.1 | no whole count | 0.9 to 23.6 | 2.8 to 30.1 | 8.1 to 41.6 |
| 50 | 0 to 7.1 | 0.4 to 10.5 | no whole count | 4.3 to 21.4 | 11.2 to 33.0 |
| 100 | 0 to 3.7 | 0.6 to 7.0 | 2.2 to 11.2 | 5.5 to 17.4 | 13.3 to 28.9 |
| 200 | 0 to 1.9 | 0.8 to 5.0 | 2.7 to 9.0 | 6.6 to 14.9 | 15.0 to 26.1 |
| 400 | 0 to 1.0 | 1.0 to 3.9 | 3.3 to 7.6 | 7.4 to 13.3 | 16.4 to 24.2 |
Each cell is the range for the true reply rate, in percent. At 20 messages, one reply and four replies give ranges that overlap from about 8 to 24 percent. At 400 messages the ranges for 5 and 10 percent barely touch, between 7.4 and 7.6 percent. The weekly outreach tally carries the formula for counts between the rows.
A fixed daily number
Alex Hormozi's book $100M Leads (2023) changes a channel that works in a set order: more volume, then one improvement at a time, then new channels. This guide follows that order further down, and before any change it holds a fixed daily number of outreach actions long enough to judge the channel.
On Leadbender the daily number is sized to the replies one person can answer the same day, which the guide to cold outreach that earns a reply works out as a ceiling.
From a revenue target to a weekly count
The weekly number of sales conversations comes from four figures a seller already has: the new revenue target, the average first-year deal, the share that moves forward at each step, and the selling weeks left in the period. Renewals and repeat orders stay out, because they come from existing accounts, not from outreach.
The company and every number in this example are made up. A two-person firm sells monthly bookkeeping to small contractors and wants $120,000 in new first-year revenue over 24 selling weeks, with an average new account worth $6,000 in the first year.
| Step | Rate used | Calculation | Answer |
|---|---|---|---|
| Deals needed | $6,000 per deal | 120,000 divided by 6,000 | 20 deals |
| Proposals needed | 40 percent of proposals sign | 20 divided by 0.40 | 50 proposals |
| Qualified conversations needed | 50 percent reach a proposal | 50 divided by 0.50 | 100 conversations |
| First conversations needed | 25 percent qualify | 100 divided by 0.25 | 400 conversations |
| Per selling week | 24 selling weeks | 400 divided by 24 | about 17 a week |
Seventeen first conversations take about 13 hours a week once preparation and notes are counted. The qualified calls and proposals that follow add about 12 more, for about 25 hours of selling before anyone looks for a new name. The founder in the example has ten hours.
| Activity | Per week | Hours each | Hours per week |
|---|---|---|---|
| First conversations | 17 | 0.75 | 12.75 |
| Qualified conversations | about 4.2 | 2 | about 8.3 |
| Proposals | about 2.1 | 2 | about 4.2 |
| Total | about 25 |
A short screen before the first call that lifts the share of first conversations that qualify from 25 to 40 percent cuts the example from 400 first conversations to 250. A longer period or a lower target cuts the 17 a week; more selling hours raise the ten. Pricing and packaging change the deal value, and this site does not teach them.
How fast a reply has to come
In March 2011 the Harvard Business Review published a study of response speed by James Oldroyd, Kristina McElheran and David Elkington. Across 1.25 million online sales leads received by 29 consumer and 13 business-to-business companies in the United States, firms that tried to reach a lead within an hour were nearly seven times as likely to qualify it as firms that waited even an hour longer, and more than 60 times as likely as firms that waited 24 hours or more (Oldroyd, McElheran and Elkington, 2011, subscription). The authors counted a lead as qualified when the firm had a meaningful conversation with a key decision maker.
The leads were inbound web inquiries, most of the companies sold to consumers, and one author, David Elkington, was chairman and chief executive of InsideSales.com, a company that sells sales software. Replies to cold outreach are a different kind of lead, and the study did not measure them.
Changing one thing at a time
In the order this guide uses, a channel first gets more volume until something breaks. Then it gets one fix at a time at the same volume, and a new channel comes only after that. The range table above sets how many messages a change needs before its effect can be told apart from chance.
Made up for this example: the bookkeeping firm sends 40 cold messages a day for two weeks, 400 in all, and gets 12 replies. It then changes only the first line and sends another 400, which bring 22 replies. The Wilson ranges for the two results, about 1.7 to 5.2 percent and about 3.7 to 8.2 percent, still overlap between 3.7 and 5.2, so the firm keeps the new line and sends two more weeks before calling the change a win.
How many open deals the pipeline holds
Little's Law, which John Little proved in 1961, says that in a stable system, over the long run, the average number of items inside equals the rate at which they arrive times the average time each one stays (Little, 1961, abstract). In a sales pipeline, open qualified opportunities equal new qualified opportunities per week times the weeks from qualification to a decision.
Made up for this example: the bookkeeping firm takes 6 weeks from qualification to a decision. At about 4.2 new qualified conversations a week, that gives about 25 open opportunities at any time. Forty open opportunities at the same arrival rate would mean deals are taking about ten weeks, not six. Fifteen, with the same six-week wait, would mean only about 2.5 new qualified conversations a week are arriving, against the 4.2 the target needs.
A daily goal counted in engaged leads
A daily goal can count results instead of hours: a number of engaged leads, meaning replies that state a problem. In the made-up plan, the firm needs about 17 first conversations a week, so the day's goal is three or four engaged leads, and the day's outreach stops at the reply ceiling even when the goal is not met.
Checking the count each week
- Copy the week's messages, replies and engaged leads from the weekly outreach tally.
- Find the week's reply rate in the range table above, or work it out with the formula on the weekly outreach tally, before judging it.
- Count first conversations held against the weekly number.
- Count open qualified opportunities against the number Little's Law gives.
- Once a month, recompute the rates from the deals that closed or died, and rerun the chain before changing the outreach.
Sources
- NIST/SEMATECH e-Handbook of Statistical Methods, section 7.2.4.1, "Confidence intervals." https://www.itl.nist.gov/div898/handbook/prc/section2/prc241.htm
- James B. Oldroyd, Kristina McElheran and David Elkington, "The Short Life of Online Sales Leads," Harvard Business Review, March 2011 (subscription). https://hbr.org/2011/03/the-short-life-of-online-sales-leads
- John D. C. Little, "A Proof for the Queuing Formula: L = lambda W," Operations Research 9(3), 1961, pages 383 to 387 (abstract). https://pubsonline.informs.org/doi/10.1287/opre.9.3.383