Cold email CTAs: the ask that gets replies
Ask for interest in something specific, not for time: a stranger has not made the decision your calendar ask assumes. After one reply, the calendar ask becomes the right one. This post is why that reverses, what to ask at every step, and what to send in the half hour after a yes.
By Kshitij Maheshwari, co-founder · Updated August 2026 · 11 min read
Ask for interest, not for time
One dataset ranked the same three asks twice, once on cold email and once inside a live deal. Between the rows the order does not shift, it reverses.
| Where the reader stands | Specific ask: a day and time | Open-ended ask: a meeting, no date | Interest ask: something specific |
|---|---|---|---|
| Cold email, nothing spent yet | 15%Asks for a decision nobody made | 13%The same decision, vaguer | 30% WinsThe decision costs almost nothing |
| Live deal, already evaluating you | 37% WinsRemoves the last chore left | 32%Leaves the chore in | 25%Reopens a settled question |
Source: Gong Labs, Devin Reed, 27 May 2020, checked August 2026. Vendor data, 304,174 emails, charts read from the source images. Gong's success definition: "a positive outcome, meaning a meeting booked in the next 10 days".
Every ask bills the reader twice
An ask carries two prices, and they move independently. Which one kills you depends on where the reader is standing when it lands.
The decision cost is what the reader has to settle before answering. The chore cost is the clicking, scrolling and typing that answering takes. Also see positive reply rate.
- •What the reader must conclude before they can answer
- •Fifteen minutes Tuesday assumes a conclusion nobody reached
- •Unpaid on a cold email, so it is the whole bill
- •Falls to nothing once they are evaluating you
- •The physical work of answering, measured in seconds
- •Trivial on a first touch: one word, typed with a thumb
- •Dominant in a live deal, where the yes is already settled
- •Naming a time deletes it, which is why precision wins late
The rest of the message, the opener, the one claim, the length, belongs to our cold email copywriting guide. This post prices the last sentence only.
A cold email dies on the decision. A live deal dies on the chore.
Why the ranking flips
Nothing about the sentence changed between those two rows. The reader did, and that is the whole mechanism.
The decision is unpaid
Time is the one thing your reader cannot make more of, so a stranger asking for it is asking for a loss. They have no basis yet for a yes, and the cheapest answer to a question you cannot answer is no answer at all.
The decision is already made
They agreed to meet weeks ago, so there is nothing left to conclude. What is left is calendar admin, and a named slot does that admin for them: they reply yes, or they reply with a better time.
Same sentence, different bill
So the specific ask more than doubles between the two rows, and the interest ask falls. Asking a buyer who is already sold whether they are interested reopens a question they had closed.
Source: the same Gong Labs analysis as the table above. The explanation of why the ranking flips is ours.
The test: have they spent anything?
The flip is useless without a way to tell which side of it you are on. Touch count is not that test. Payment is.
An illustrative walkthrough of the method, not a specific client result. We report real numbers only when they are real. The example copy is invented, and no reply rate is attached to it.
Count touches and escalate
Circling back one more time. Do you have 20 minutes Thursday or Friday to walk through this?
- ✕Six sends is your effort, not their investment
- ✕They concluded nothing, so the decision is still unpaid
- ✕Offers a choice, which is a second decision
Hold the rung until they spend
Want the benchmark we pulled for your segment? Happy to send it and leave it there.
- ✓Same size ask, because nothing was earned yet
- ✓Answerable in one word, from a phone
- ✓Names the thing that arrives if they say yes
Have they replied, answered a question, or asked for something you then sent? If not, you are cold on touch six exactly as you were on touch one. Touches are not trust.
The best closer on a team is often its worst cold emailer, and it is not a skill gap. They carry an ask that works all day in rooms where the decision is already paid, then send it into an inbox where nobody has paid anything.
A real event does not buy you a bigger ask either. It buys a reason the smallest ask is worth answering, which is what the trigger event play is built on.
How high the ladder goes
Asks come in sizes. You move up one rung each time the reader spends something, and the top rung on cold is not the interest ask.
The smallest rung on the ladder
Are you the one who owns this now? A correction costs nothing to give and opens the ledger. No dataset measures it, so treat it as practice: use it when the offer is not ready.
Good, and no longer the ceiling
The winner on cold email in the table above, and still the safe default. Five years later, on 30MPC and Gong's 85 million cold emails, it scores +7% on reply rate. Buyers see so many cold emails now that even the best ask moves the number only a little.
The top rung on a cold email
On the same 85 million emails, an offer scores +28% on reply rate and a meeting request -44%. What a stranger would actually want sent is the outbound offer question, and it comes first.
One rung after a yes, not one touch after a send
In the Gong dataset the calendar ask lands at 37% inside a live deal, and less than half that on cold email. Seniority raises the bill rather than the volume: Gong's January 2026 analysis of executive sales cycles puts C-level readers 30.2% less likely to reply than non-executives.
One rung per payment, and a mechanism not a lift
A small yes does make the next ask easier, which is why the ladder works at all. It is a mechanism you can use, not a lift you can bank: no cold email dataset prices it, so it never belongs in a forecast. Move one rung, then wait for the next payment.
The ask at every step of a sequence
Five positions, five asks. Cadence, spacing and step count belong to the outbound sequences guide. This is only the last line.
| Where you are | What to ask for | The basis |
|---|---|---|
| Touch one | Interest in one named thing, or the thing itself. | The best cold ask Gong measured. An offer scores +28% on reply rate across 85 million emails. |
| Touch two, no reply | The same ask, shorter, from a different angle. | Nothing was spent, so nothing was earned. Escalating here is the common error. |
| The bump after silence | One explicit question, never a reminder. | Explicit CTA language scores +27% on reply rate. "Quick chat" scores -2%, which buys nothing. |
| The last touch | Permission to stop. | Breakup language scores +89%, the highest in the set. Nobody pays a cost to answer it honestly. |
| After a yes | A specific day and time. | The decision is paid now, and the specific ask wins by a distance. The only place a calendar belongs. |
Sources: 30 Minutes to President's Club with Gong, The Ultimate Cold Email Data Report, written by Jason Bay, 2025, vendor data, 85 million cold emails measured on reply rate; charts read from the source PDF. Gong Labs, 304,174 emails, as above.
Our sequences guide already tells you to change the ask rather than the volume, and to end a sequence on purpose. This table is that rule written out per step.
Want the ask written and the replies worked, by the two of us?
Book a Fit CheckWhat to do with a yes
Top reps turn replies into meetings at 29.6% against 15.3% for the average, on that same 85 million email set. Half the gap sits after the reply.
-
1
Send the thing, and only the thing
The reply bought exactly one item. Put that item at the top of the message with nothing bolted onto it.
-
2
Same day, ideally the same hour
A reply is a window of attention on somebody who has just decided you are worth a minute. Nothing in your sequence design reopens that window.
-
3
Put one ask underneath it, one rung up
After an interest ask, that is a specific day and time. The decision is paid now, so precision reads as a courtesy rather than a cost.
-
4
Count what you attached
A deck, a link, a case study and a thirty minute request is four rungs at once, billed to a reader who paid for one.
Before you send an ask, answer one question: what exactly arrives if they say yes? If the honest answer is "a conversation", the ask has nothing behind it, and the reply you get back is the one nobody has a plan for.
Eight ways an ask fails
Eight failures, four shapes. Each pair breaks for the same reason, which is why fixing one usually fixes its twin.
A second ask is a second decision. The fake-choice close is the same error in better manners: Tuesday or Thursday still assumes the yes you never got.
An ask buried mid-paragraph and a bump with no question in it fail identically. If there is nothing to answer, nothing gets answered.
A demo request dressed as curiosity, and "worth a quick chat?", both still ask for the meeting. "Quick chat" scores -2% on reply rate: it buys nothing.
"Let me know if you would like to learn more" and "any thoughts?" name nothing. Gong reads "thoughts" as a word that can lift replies while lowering meetings booked.
The calendar link in touch one
It is the calendar ask in a costume, and it is also a click a stranger has no reason to make. Nobody has published data on links in first touches, but complaints land on your domain rather than on the campaign, which is why the copywriting guide keeps links out of a first email.
What the numbers do not say
Two things to hold onto before that number goes anywhere near a forecast.
"Interest CTAs convert at 30%. Change your last line and book three times the meetings."
- ✕Reads a within-population ranking as a benchmark
- ✕Ignores whose emails were measured, and when
- ✕Sells a rewrite as a pipeline forecast
"Across 304,174 emails, one ask outperformed the two others Gong measured, in Gong's own population."
- ✓A ranking of three asks, not a rate you should expect
- ✓The same vendor's 85 million email set: 3 meetings per 1,000 sends for the average rep, 23 for the top 10%
- ✓Two orders of magnitude apart, so the level does not travel
The second thing: a reply is not a meeting, and the two can move in opposite directions. Pick the outcome you want before you pick the ask.
And a great ask does not rescue a missing offer. If nothing exists that a stranger would want sent to them, no wording produces one, which is what the outbound offer guide is for.
Key takeaways
Four lines that decide the last sentence of every cold email you send.
- 1 Ask for interest on cold, for a time after a yes.
- 2 Payment moves the ask up a rung, never touch count.
- 3 One ask a message, with a named thing behind the yes.
- 4 Deliver first, same day, then move up exactly one rung.
Questions founders ask
What is the best call to action for a cold email?
When is it right to ask for a specific day and time?
Why does the best CTA change once a deal is live?
Is "open to a quick chat?" a soft CTA?
What should the follow-up after silence ask for?
They replied "sure, send it". Now what?
Co-founder of Real Good GTM. He has been the first business hire and Chief of Staff at seed-stage B2B startups, building outbound pipeline before any playbook existed. This post comes from handling the replies himself, where the half hour after a yes decides more than the wording of the first email did.
Connect on LinkedInThe rest of the message
The ask is the last line. These three cover the reason, the opening and the shape.
Cold email copywriting
The whole message: the five parts, the one claim you get, and what the length rules actually measured.
Read the guideThe reasons to reply library
Why a stranger answers at all, sorted and named. The reason underneath the request.
Read the postCold email opening lines
The first line decides whether the last one is read. The patterns that earn line two.
Read the postWant an ask that earns replies you can work?
Book a fit check. We'll look at what you could credibly offer a stranger, write the ask that fits it, and tell you straight if outbound is not the right motion for you yet.
Book a Fit CheckNo hard sell. No fake numbers. Real good work speaks for itself.