Pillar 1: The Virtual SDR and outbound model

“Virtual SDR” gets used loosely enough that it’s worth being specific about what the role actually does, day to day, once a campaign is live.
What it is not
It isn’t a pure-automation tool left to run unsupervised, and it isn’t a black-box agency that disappears behind a monthly report. Both extremes fail the same way, one for lack of judgment, the other for lack of visibility. A Virtual SDR sits deliberately in between: AI carries the sourcing and the sequencing, a person carries every judgment call, and the client can see exactly which of the two is doing what at any point in the week.
How we approach it differently
Most providers in this category pick one extreme and stay there. A software tool sells full automation and calls it a Virtual SDR because the label tests well, then goes quiet the moment a prospect asks a real question. An agency sells an offshore team and calls it AI-powered because the phrase is fashionable, when in practice someone is copy-pasting from a script with no real targeting behind it. We built ours the other way round. The same person owns a client’s list from the first message through to the twentieth booked meeting, so the judgment calls stay with someone who actually knows the account, not whoever’s free that day. AI only ever touches the repeatable steps, sourcing and scoring and sequencing, never the reply. And it’s visible weekly, not summarised into a monthly report the client has to take on faith. That’s also why the model scales the way it does, one focused rep to start, a Full Squad once the targeting and messaging are proven, rather than a large team thrown at the problem from day one.
Getting started: the ICP and the first list
Nothing goes out before the target is defined properly: industry, seniority, company size, geography, timing signal, buying signal, budget. A human sets that target. From there, AI pulls and scores a first batch of prospects against it, usually fifty to a hundred names, not the full list, so the targeting can be checked against real replies before it scales.
The daily cadence
Once a list is approved, outreach runs on a fixed sequence: a connection request, then a short, genuinely curious message a few days later, then a light follow-up, then, only once two decent touches have landed, a message that finally mentions the product. Every account stays inside safe sending limits, so the pace is steady rather than aggressive, a small number of new touches added each day rather than a batch sent all at once.
When a reply comes in
This is the part that doesn’t run on autopilot. A reply gets read in full, against the thread and the prospect’s score, and sorted by what it actually says: interested, asking for proof, raising an objection, or not interested. A person decides how to respond, or approves a drafted response before it goes out. Objections in particular need judgment. A templated reply to “we already have someone for that” reads as exactly what it is, and a prospect who catches that stops replying to anything after.
A week in the life
Monday sets the theme and reviews what came in over the weekend. Tuesday and Wednesday are the heaviest send days, new connections and first messages going out inside the daily limit. Thursday is largely reply-handling and follow-ups on threads that have gone quiet for a few days. Friday is the review: acceptance rate, reply rate, and which messages earned a real conversation versus a polite one, feeding directly into the following week’s targeting and copy.
Booking and beyond
Once a meeting’s confirmed, the handoff to the client’s team is deliberately plain: a clean brief on who the prospect is and what’s already been discussed, no over-selling. Every week, the whole system gets reviewed by segment: which messages are converting, which accounts are slowing down, what’s worth adjusting before the next batch goes out.
Why the split holds up
AI carries the repetitive load, sourcing, scoring, sequencing, so a person’s time goes to the parts where judgment actually matters. Well-run campaigns typically settle into eight to twenty-five-plus booked meetings a month, usually from month two once targeting and messaging are dialled in. That timeline isn’t a limitation of the system. It’s what building a list and a reputation on a channel actually takes, and it’s the same operating rhythm behind 87 meetings in 90 days on a single channel, LinkedIn only, six SDRs, nothing else running in parallel.
Leadsourcing builds and runs Virtual SDR campaigns for B2B teams on LinkedIn, one channel,
run properly, with a real person behind every reply. If you want to see what that looks like
against your own list, book a call and we’ll walk through it.
