# How to automate LinkedIn replies for sales
> Choose between hand-off, interest-based and booking replies on LinkedIn, and learn what to write in the playbook so the AI sounds like your own voice.
- HTML: https://getkinlead.com/automate-linkedin-reply-handling-modes
- Markdown: https://getkinlead.com/automate-linkedin-reply-handling-modes.md

**Short answer:** Automating LinkedIn replies comes down to one decision: who answers the first reply. Fixed-scope offers can go straight to a booking link, while custom or high-value work needs hand-off mode so a person takes over the moment someone replies.

## The three ways to handle a LinkedIn reply

Every campaign that gets a reply has to decide who answers it. There are three settings, and each one hands the conversation back to you at a different point. Choose per campaign rather than once for the whole account, because the right answer changes with the list.

Hand-off mode stops the automation at the first reply. The thread lands in your inbox and you write the answer yourself. Nothing else goes out until you do.

Interest-based automation lets the AI handle the opening exchanges and step aside once the person says something that sounds like a buying signal: they describe the problem, ask about scope, mention a timeline.

Booking automation keeps the AI in the thread until a meeting sits on your calendar. The first time you speak to the person is the call itself.

**Pushed the meeting too early**

- Them: We run four systems and one is a homegrown ERP nobody understands.
- AI: That sounds great! Grab a time here and we can go through it.

**Answered the detail first**

- Them: We run four systems and one is a homegrown ERP nobody understands.
- Draft: Homegrown systems come up a lot. The first call is usually about what has to stay and what can be replaced. Does Tuesday at 10 work, or do you prefer later in the week?

## Matching the reply mode to your offer

The question to ask is what a misunderstanding costs you. If someone can read your offer on a page and know whether they want it, booking automation will not hurt. If they need to describe their setup before your offer makes sense, the AI has to hear that description before it pushes a meeting.

Price matters less than variation. A cheap product with ten configurations is harder to automate than an expensive service that looks the same for every client.

Look at your last ten closed deals and count how many needed a question from you before the buyer understood the offer. If that number is low, book directly. If it is most of them, use hand-off.

Running two campaigns with different modes beats writing one prompt that switches behaviour halfway through a thread. Split by segment: enterprise titles in one, smaller accounts in the other.

| | Offer type | Reply mode | What the AI does |
| --- | --- | --- | --- |
| Low price, fixed scope | Booking | Answers from your playbook, then sends the scheduling link |
| Mid price, repeatable service | Interest-based | Qualifies against your bar, hands over when scope or pricing comes up |
| Custom scope, long cycles | Hand-off | Nothing. The reply waits for you |

## When hand-off mode is the right call

Hand-off earns its place when the reply carries information you need to read yourself. A named system, a regulation, a reorganisation, a previous vendor. Those details change what you say next, and the AI has never seen them.

It also protects you when your offer changes per client. A consultant working across healthcare and fintech answers a hospital operations director and a fintech founder differently. The AI only knows what you wrote in the playbook, and you wrote it for the average case.

The cost is time and speed. Turn on reply notifications, check the inbox before lunch, and answer the same working day. A hand-off that sits for three days reads worse than a mediocre automated reply.

Keep hand-off campaigns small. You are the bottleneck, so cap the number of leads in flight at whatever you can actually answer in a day.

- **them:** Do you work with companies over 500 people in the EU?
- **you:** Yes. We finished a project for a 600 person firm in Berlin last quarter. Most of the work was GDPR around the HR module. Is that the part you are worried about, or is it something else in the stack?
- **them:** GDPR is fine, it's the payroll data sitting in three places. That's the mess.

## Setting up an interest-based flow

Interest-based automation does one job: decide whether a person is worth your afternoon. Write that definition down before you switch the mode on.

Vague definitions produce vague handovers. Interested means nothing. They described the problem we fix and asked about scope or timing is something the AI can actually check against a message.

The hand-off trigger is the part people skip. Without it the AI keeps chatting politely and you hear about the warm lead a week later, after they have booked something else.

Set the tone in the same place you set the bar. Two or three sentences, no exclamation marks, no more than three sentences per reply. Short rules get followed.

1. **Write the qualifying bar in one sentence.** Say it out loud. If it takes two breaths, it is too long. They run outbound for a team of three or more and say reply rates dropped this quarter.
2. **Answer the five objections you hear most.** Exact wording, your words, including the price one. This becomes the playbook, and it stops the AI improvising around your pricing or your delivery timeline.
3. **List the hand-over phrases.** Pricing, contract, security review, a named competitor, can I talk to a person. When one appears, the AI stops and you get notified.
4. **Read the first twenty AI replies yourself.** You will spot the phrase it keeps getting wrong faster in a live thread than in a settings page, and you can fix the playbook the same afternoon.

## What full booking automation gets wrong

Two things break. The first is trust. People work out they are talking to a script, usually at the point where their specific detail gets ignored and the reply repeats the pitch in slightly different words.

The second is your calendar. A bot that books anyone who says yes fills your week with calls that were never going to close. Count the real cost of a bad booking: prep, the call, the follow-up, and the slot a good prospect could have taken.

Tell the AI how long the call is and what happens on it. A fifteen minute walkthrough is an easy yes. A sixty minute discovery session needs a person on the other end deciding whether to give up an hour.

Booking automation holds up when the offer is fixed, the price is roughly public, and the call is a demo. It falls apart when the call is a discovery conversation, because discovery needs someone listening and following up on what they heard.

> Give the AI one scheduling link and one sentence about what the call covers. If the wrong people keep booking, tighten the qualifying bar. A longer script will not fix a loose bar.

## Testing a reply mode before you trust it

Run the mode on a small batch first. Twenty leads in one campaign, watched by hand, tells you more than a week of reading settings pages.

Send yourself the messages too. Put your own profile into a small campaign and reply as a difficult prospect: ask the price, dodge the question, name a competitor, go quiet for two days.

Score three things per reply. Did it answer the actual question, did it stay in your tone, and did it stop when the person showed real interest. If two of those fail, the playbook is the problem, not the mode.

```
Who we are: [one line]
What we sell: [one line, no adjectives]
Qualifying bar: the person runs [X] and says [problem] is costing them [Y].
Hand over now when they mention: pricing, contract, security review, a named competitor, or asking for a person.
Answers:
- How much does it cost? -> [your exact wording]
- Do you work with [tool or setup]? -> [your exact wording]
- How long does it take? -> [your exact wording]
Tone: short sentences, no exclamation marks, three sentences maximum per reply, British spelling.
```

## Doing this with Kinlead

Kinlead sets the reply mode per campaign. You can hand the conversation to yourself at the first reply, let the AI reply until there is real interest, or let the AI reply until a meeting is booked using your scheduling link.

Messages and follow-ups are written from the campaign's goal, tone and an optional sales playbook, which is your own sales script. Queued messages can be read and edited on the Actions page before they send, and any campaign can be paused. Replies from every campaign land in one [unified inbox](/features/unified-inbox) classified by intent, interested replies first, and you get an email when someone replies.

If you want to read every reply before it goes out, use hand-off. In the two AI reply modes the replies send on their own, with no approval step.

Pro has a [minimum booking guarantee](/minimum-booking-guarantee): after a two week warm-up, a week with fewer than three qualifying bookings, with at least one agent active all week, can be claimed as a refund minus transaction charges of about 3.5% to 5%.

## Related

- [Minimum booking guarantee](https://getkinlead.com/minimum-booking-guarantee.md)
- [AI LinkedIn outreach](https://getkinlead.com/features/ai-linkedin-outreach.md)
- [Unified inbox](https://getkinlead.com/features/unified-inbox.md)

## Common questions

**Does AI automation work for high-ticket B2B sales?**

Yes, with hand-off mode or a strict interest-based bar. Deals that need research per account fall apart when a bot answers every message on its own.

**How do I stop the AI making things up in my LinkedIn messages?**

Give it a playbook with your exact wording for price, timeline and delivery questions. When the AI has approved answers to lean on, it has less room to invent facts about your product.

**What happens when a lead asks something the AI cannot answer?**

In the interest-based and booking modes the AI answers from your playbook. Add the phrases that should always stop it, such as pricing, contract or a request to speak to a person, so those threads come to you.

**How many leads should I put in a test run?**

Enough to see a few replies and few enough to read every one. A small batch watched closely beats a large one you only check on Friday.

[Create a Kinlead account](https://getkinlead.com/signup) or [book a call](https://getkinlead.com/demo) to talk through pricing and setup.
