InMail vs connection request vs email: which first touch for which buyer

Use a connection request as the default first touch: it is free and starts a relationship. Use an InMail for a person you are not connected to who matters enough to spend a credit on, or whose Open Profile makes it free. Use email when you have a verified address and the person is not active on LinkedIn. We publish no InMail result, because we have too few sends to measure.

Three ways to start a cold conversation, each with a different cost and a different way to fail. The posts that compare two of them are common. This one gives a decision rule for all three, and says plainly where our evidence stops.

What are the three first touches?

A connection request. An invitation to connect, with or without a short note. It is free. If accepted, you can message the person. If ignored, you have spent one of your limited invitations, and a pile of ignored or pending invitations is among the reasons LinkedIn lists for a restriction.

An InMail. A private message to someone you are not connected to. It normally spends an InMail credit that comes with a paid LinkedIn plan, and it is free when the recipient has an Open Profile. The message arrives without the person accepting anything. See the glossary entries on LinkedIn InMail and Open Profile.

An email. A message to a verified address. It needs an address, a sender setup, and respect for the rules of email, and it can carry more text.

How do they compare?

What is the decision rule?

Ask five questions in order.

1. Do you have a verified email and does the person seem inactive on LinkedIn? If yes, email first. Many senior people rarely open LinkedIn.
2. Is the person an Open Profile member? If yes and the lead matters, an InMail costs nothing.
3. Is the person high value and hard to reach? If yes and you have credits, an InMail is worth a credit.
4. Is the list large and the value per lead modest? A connection request.
5. None of the above? A connection request, with no note, and an observation opener after it is accepted.

On our own account, accepted connection requests did well: 41% of 68,818 were accepted. The first message after acceptance mattered: an observation opener beat a question by 10 to 15 points. See observation openers beat questions.

What do we know about InMail?

Little, and we say so. Sellinger's InMail first touch is live for all accounts, but we have too few sends to publish a result, and we make no claim that it beats a connection request. What we can describe is how it works: the Sellinger agent opens a lead it is not connected to with an InMail, free for an Open Profile member and one Premium credit otherwise, and falls back to a connection request when no credit or daily slot is left. The default is 10 a day per account, with a maximum of 50. The account needs LinkedIn Premium or Sales Navigator. See the InMail product page and how the agent decides.

The credit planner helps you work out how far your credits go, with your own estimate of the Open Profile share.

What do we know about email?

More. The Shelly installer programme emailed 3,940 installers in 111 countries, and 13.6% replied in the June to August waves, against 3.3% for the first email alone. That was a listening programme to an existing base, not a cold sales campaign, but it shows that email with follow ups can reach people who would never accept an invitation. Email defaults to 50 sends a day per mailbox, with a suppression check before every send. Read the minimum setup for cold email first.

When should you use more than one?

Often. A multichannel sequence can start on LinkedIn and move to email for people who stay silent. Sellinger's defaults move a lead to email after four LinkedIn follow ups, and every value can be changed per campaign. See a multichannel sequence that works. A reply on any channel should stop the others.

How do you test your own rule?

Take 300 people with both a LinkedIn profile and a verified email. Split them into three groups of 100 by random draw: connection request, InMail (if you have credits), email. Use the same offer, and comparable senders. After four weeks count replies and meetings per person messaged. Three groups of 100 are small, so look for large gaps. Whatever you learn is worth more than any vendor's claim, including ours.

How long does each first touch take to reach a list?

Daily limits decide how fast a list can be worked, so do the arithmetic before you plan a launch. These are worked numbers from the Sellinger defaults, with a made up list of 600 people.

The connection request line follows the weekly cap of 150 invitations. Four weeks is the earliest, because the weekly cap binds first. The email line assumes no warm up. With the optional warm up, which starts at 5 a day and doubles every 3 days up to the limit, the first two weeks are slower. Every limit can be set per account. See daily limits and pacing.

The point of the table is to match the touch to the list. A list of 600 where each person matters belongs in a slower, richer touch. A list of 6,000 cannot be reached by InMail at the default pace at all, and needs connection requests and email.

A made up example: three leads, three touches

These three leads are invented to show the rule in use.

• Lead 1: a founder with an Open Profile and a verified email. An InMail costs nothing, so send it, and keep the email for a follow up if there is no reply.
• Lead 2: a finance director at a mid sized company, no Open Profile, a verified address, and a profile that looks dormant. Email first. A credit spent on a dormant profile is a credit wasted.
• Lead 3: a sales manager with an active profile and no verified email. Connection request, with no note, then an observation opener once accepted.

Same decision rule, three different answers, because the data differs.

Which mistakes show up most?

• Spending InMail credits on a list of low value leads. Credits are scarce. Keep them for people who matter.
• Sending a connection request note that is a pitch. The invitation is a door, not the sales call.
• Running all three at once on the same person. Pick one first touch and let the others follow only on silence.
• Judging a touch on a week of data. Replies come over days, and acceptances come late.

A checklist

• [ ] The default first touch is a connection request.
• [ ] I know my InMail credits and my daily InMail limit.
• [ ] InMails go to people who matter, or Open Profile members.
• [ ] Email is used where I have a verified address and the person is quiet on LinkedIn.
• [ ] A reply on any channel stops the others.
• [ ] I plan a split test before I trust a rule.