Building an outreach continuity plan for when accounts go down

Design for the outage instead of against it

Assume the account goes down and build backwards from there. LinkedIn's User Agreement asks members not to share accounts, accounts can be restricted for a range of reasons, and even a well-run profile can be caught by a verification prompt or a wave of enforcement. Nobody can remove that possibility. What you control is what an outage costs you, and that number is set by decisions taken months earlier.

Most teams that lose a quarter to a restriction did not lose it to the restriction. They lost it because the conversation history, the contact details and the follow-up schedule existed nowhere except inside the account. Understanding why accounts get restricted reduces the frequency; continuity planning reduces the cost, and only the second is fully in your hands.

Start with the four-hour test

Ask one question of your current setup: if the main account went dark right now, how many live conversations could you continue within four hours? Count them honestly. A conversation you can continue is one where you hold the person's name, company, an email address or phone number, and a record of what was last said. Anything else is a conversation you would have to abandon and restart cold.

The arithmetic is your own and it is unflattering the first time. If you have 180 live conversations and hold an email address for 40 of them, you could continue 40 — about 22% of your pipeline survives the outage and the rest goes back to square one. Run that calculation before you read the rest of this page, because it tells you which of the following sections is actually urgent for you.

Continuity is not a tool you buy. It is a number you can quote about your own pipeline before anything goes wrong.

Mirror the data out on a schedule

Daily, into your CRM, automatically if possible and manually if not. The goal is that the account holds a working copy of your outreach rather than the only copy. A weekly export is better than nothing, but restrictions do not wait for Fridays.

1. Message threads with timestamps, so a colleague picking up the conversation can see what was promised and when.
2. The accepted-connection list, with company and title, exported on a fixed cadence.
3. Pending invitations, which are your live top-of-funnel and the first thing to vanish from view.
4. Any notes and tags — if these only exist inside a Sales Navigator list, they are tied to a seat and not to you.
5. Saved search definitions, so the targeting logic is reproducible on a different account.
6. A per-conversation next step and date, held in the CRM rather than in someone's memory or the account's inbox.

If you are running a rented or managed profile, put the export cadence in the agreement rather than trusting a process. Export should be an ongoing right during the term with a stated cadence, not a handover promised at the end — an end-of-term handover is worth nothing in an incident that happens in month two.

Give every live conversation a second channel before you need it

Capture an alternative contact point at the earliest natural moment in each conversation, not at the point of crisis. Asking for an email address after your account has gone dark is impossible, because you no longer have a channel in which to ask.

The discipline is the same one good sales teams already apply for a different reason: conversations that stay locked inside a social inbox convert worse than ones that move to email and calendar. Continuity is a side benefit of a habit that pays for itself anyway. Where we run appointment setting, the handoff to email and calendar happens as early as the conversation allows, and the account stops being a single point of failure at that moment.

Cap the share of pipeline any single account carries

Decide the maximum percentage of live conversations one account may hold, then route new volume to respect it. Without a cap, volume drifts to whichever account is performing best, which quietly concentrates the entire programme onto the one profile whose loss would hurt most.

Do the arithmetic with your own numbers. If you have 240 live conversations spread across three accounts and one of them carries 150, that account holds 62% of your pipeline and a restriction there is a 62% outage. Setting a ceiling of, say, one third means the same event costs you a third — and the other two thirds keep moving while recovery runs. The ceiling you pick is a business judgement about how much of a quarter you are willing to put on one identity.

Capacity has to be staged rather than created on the day you need it. A brand-new profile cannot absorb a restricted account's volume immediately without behaving in exactly the way that attracts attention. Keep the standby identity warm and lightly active so it is ready, and size the number of accounts to the volume rather than the other way round — the sizing logic is in how many accounts to run.

The redundancy trap: correlated accounts are not redundancy

Spinning up several accounts on one machine, one network and one behaviour pattern does not spread risk — it synchronises it. This is the mistake that turns a plan into a worse outage, because the accounts you were relying on as backup go down in the same wave as the original.

• Same device and browser profile for every account, so the sessions look like one operator wearing hats.
• Same network egress, so everything traces to one place.
• Identical message copy across accounts, which makes a pattern easy to recognise across profiles.
• Identical daily timing and volume, which does the same thing from a different angle.
• Accounts created in one burst, with the same recovery email domain, so a single review reaches all of them.
• One operator running all of them at once, which caps how human the pacing can genuinely be.

Genuine independence means different operators, different networks, different devices, different message angles and staggered activity — plus profiles with real, distinct histories rather than freshly minted ones. That is the practical argument for separately maintained identities with real histories over a pile of new sign-ups. Redundancy that fails together is a cost, not a control.

The restart runbook

Write this down now, one page, and store it outside the account. During an incident nobody invents a good process; they execute whatever was already written or they improvise badly.

1. Hour 0: stop all sending on the affected account. Do not attempt repeated logins or fresh verification attempts in quick succession.
2. Hour 1: confirm you already hold the export. If you do not, this is the moment you discover the gap — note it and fix it permanently afterwards.
3. Hour 2: message every live conversation on their second channel, briefly and without drama. A short note that you are following up by email is enough.
4. Day 1: begin the recovery attempt through LinkedIn's own appeal process, with one named owner of the task and a reporting cadence. What that involves is set out in account recovery.
5. Day 1-2: route new volume onto the staged standby identity at reduced pace, not at the restricted account's previous volume.
6. Week 1: review what concentrated the pipeline onto that account in the first place and adjust the cap.
7. Week 2: close the loop on the conversations that had no second channel — the ones you had to abandon are the measurement of what the plan still misses.

If you would rather not build the staging capacity in-house, that is the part we handle as a service: managed identities that are already warm, a replacement path when one drops, and a profile live in 48 hours rather than starting from zero mid-incident.

Key takeaways

• Run the four-hour test: how many live conversations could you continue if the account went dark now?
• Mirror threads, connections, pending invitations and next steps into your CRM daily.
• Capture a second channel at first reply, not at the moment of crisis.
• Set a ceiling on the share of live pipeline any one account may hold, and route volume to respect it.
• Accounts on one device, network and copy pattern fail together — that is concentration, not backup.

Frequently asked questions

How many accounts do I need for real redundancy?

The count matters less than independence and staging. Three genuinely separate identities with different operators, networks and message angles are more resilient than six created the same week on one machine. Size the number to your volume and your pipeline concentration cap rather than picking a number first.

What should I export from LinkedIn, and how often?

Message threads with timestamps, the accepted-connection list, pending invitations, any notes or tags, and saved search definitions — daily if you can automate it, weekly at the absolute minimum. The test is whether a colleague could pick up every live conversation from the CRM alone.

What do I say to prospects when an account goes down?

Keep it short and unremarkable: you are following up by email, here is where the conversation left off, here is the next step. Do not explain the account problem. Most people will not notice or care, and the ones who do will judge you on whether the follow-up was competent.

Can a replacement account simply take over the restricted account's volume?

Not immediately. A new or newly warmed identity sending at a restricted account's previous volume behaves in exactly the way that draws attention, which risks a second outage on top of the first. Ramp the standby account up over time, and keep it lightly active before you need it so the ramp starts from somewhere.

Is it worth planning for this if we have never lost an account?

The plan costs a few hours of CRM configuration and one habit change in how reps ask for email addresses. The alternative cost is measured in a quarter's pipeline. Teams generally build the plan after the first incident; there is no reason it cannot be built before.