One conversation across LinkedIn, email and WhatsApp

Multichannel outreach fails when each channel keeps its own memory. The prospect who answered on WhatsApp then gets a LinkedIn follow up asking the same question. The fix is one record per person that holds every message on every channel, read by the sender before each send. Sellinger's built in CRM keeps the full conversation history across channels for that reason.

Running LinkedIn, email and WhatsApp together is common advice, because different people answer on different channels. The advice skips the hard part. Three channels mean three places where a reply can land, three sets of settings and three chances to forget what was said. This post shows what goes wrong without shared memory, walks through an example, and sets out what a unified inbox has to do. For choosing the order of channels, see which channel first and how to build the sequence.

What goes wrong when each channel has its own memory?

Five failures repeat. Each is easy to cause and hard to see from inside one channel.

None of these needs a bad tool. They need only two systems that do not talk to each other, which is the default when each channel has its own app.

What does it look like in practice?

A made up example. The names and details are invented to show the mechanics.

Dana runs operations at a small logistics company. A sender is working an outbound campaign.

1. Monday, LinkedIn. Dana accepts an invitation and replies: "Interesting, but not until the new year. Email me in January."
2. Wednesday, email. The email sequence, started on a separate tool, sends its first step: "Quick question about your Q4 routes."
3. Friday, WhatsApp. The same company's number is in the data, so a WhatsApp message goes out: "Hi Dana, did you get a chance to look at this?"

Dana asked once, politely, for January. She received two messages that ignored it. Nobody was careless. Each channel did its job and none could see the other.

Now with one record. Monday's reply is attached to Dana's contact with the note "back in January". Before any send the sender reads the record, sees the date, and holds the other channels until then. In January the follow up can say "You mentioned January on LinkedIn" and the channel can be whichever she answers fastest. The only change is where the memory lives.

What does a unified inbox have to do?

The label "unified inbox" covers a lot. These are the capabilities that matter.

• One person, one record. Match the same person across channels, by profile address, email and phone number, so a LinkedIn reply and an email reply land on one contact.
• Time ordered history. Every message, both directions, with channel and date, readable in one view.
• Read before write. The sender, human or agent, must read the record before drafting. A history nobody reads is a log.
• Stop rules that cross channels. An opt out on one channel must stop the others.
• Handover. A person can take over without losing the thread, and the agent can hand over when it is out of its depth.
• Context for tools. Calendar bookings, notes and deal stages in the same place, so the next message knows a meeting is already booked.

Our agentic CRM explainer covers the CRM side, and the best CRM for WhatsApp, LinkedIn and email outreach compares options. The WhatsApp side of sharing an inbox is in shared inbox for WhatsApp.

How does Sellinger handle it?

Facts only, from our fact register. Sellinger has a built in CRM with contacts, deals and the full conversation history across channels, and a two way sync with HubSpot. The channels it works on are LinkedIn, email, WhatsApp, Instagram and live chat. For the agent, the work is the same on each: it researches the prospect, writes the first message and every follow up, answers the replies and books the meeting on your calendar.

Several behaviours matter for the failures above:

• Opt outs stop the agent. When a contact replies with an opt out such as STOP or unsubscribe, the contact goes on the team's suppression list, the agent stops and the lead is closed out. A weak or unclear phrase never writes the permanent record: the agent stops and a person decides. Before every send on email, LinkedIn and WhatsApp the platform checks the suppression list and refuses the send when the contact is on it or when the check cannot be completed. (We found no such send check for Instagram direct messages, so we do not claim it there.)
• Absence notices are not replies. An out of office does not count as a reply, does not use up a follow up step, and the lead is contacted again 7 days after the return date given, or after 21 days when no date can be read.
• Machine mail is not a human. Helpdesk autoreplies, no reply senders and surveys never count as a human reply, and the lead goes to a person with a named reason.
• Follow ups move across channels. By default LinkedIn gets 4 follow ups before the lead moves to email, email gets 5 before WhatsApp, WhatsApp 4 before email, and Instagram 4 before email. Gaps start at 5 days on LinkedIn and 1 to 3 days on email and WhatsApp. Every value can be changed per campaign.
• People stay in control. A campaign can run in approval mode, where every message waits for a person to approve it, or send automatically. A person can take over any conversation, and the agent stops on its own when it cannot answer.

WhatsApp is limited by WhatsApp itself, so first messages run at a small daily volume, and Sellinger also answers people who write first. Instagram works the inbox: it answers direct messages, comments and story mentions.

How do you set up a single thread?

1. Decide the identity key. Choose how people are matched (LinkedIn profile, email, phone). Import with all that you have.
2. Connect each channel to one workspace. Not one tool per channel.
3. Pick the channel order for the first message. The sequence guides above help.
4. Set stop rules. One suppression list, shared. Decide what counts as a no.
5. Run in approval mode first. Read how the agent uses earlier messages before it sends alone.
6. Add the calendar. Cal.com and Calendly are live, so booked meetings show on the contact.
7. Sync your CRM if you have one. HubSpot is two way.

How do you check that it works?

Run a short audit on ten contacts who replied on more than one channel.

• Does each contact show every message in time order?
• Did any sequence send after a refusal or a booked meeting?
• Did the next message reflect what was said before?
• Could a person take over and understand the thread in a minute?

If one answer is no, fix it before adding volume. The reply rate will not warn you, because a duplicated message still counts as a message sent. See AI SDR metrics to track for what to watch instead.

What can't we promise?

We cannot promise that the same person is matched across channels every time. Matching depends on the identifiers you have, and two profiles with no shared field cannot be joined. The example above is invented. For the way channels depend on each other in a sequence, read how long a multichannel sequence should be.