Field Notes 31 May 2026 8 min read
LinkedIn outreach in a small market: what the numbers actually look like
NZ and Australian B2B networks are small enough that one bad campaign can follow you.
NZ and Australian B2B networks are small enough that one bad campaign can follow you.
The LinkedIn playbooks written for US markets don’t travel well to New Zealand and Australia. Most of them are built around assumptions that break down the moment you work in a market where everyone knows everyone, where the decision-maker you want to reach is also a friend of someone you met at a conference last year, and where a badly targeted campaign doesn’t just produce zero responses, it produces a reputation.
NZ has approximately 1.5 million LinkedIn users. Australia has around 6 to 7 million. The US has north of 200 million. That scale difference isn’t just a number. It changes how quickly you can burn a network, how much weight each individual connection carries, and how far a badly executed campaign travels through a professional community.
Working with B2B clients across Auckland, Wellington, and Sydney over the past twelve months, the clearest thing we’ve learned is that LinkedIn outreach in small markets requires a different logic than volume-first playbooks assume. The good news is that small markets also offer something the US doesn’t: genuine proximity. Decision-makers are reachable. Networks overlap in ways that can be used thoughtfully. The right message to the right person at the right moment lands differently here than it does in a market where you’re one of fifty outreach attempts arriving that same week.
The confusion in the market right now is between AI-assisted outreach and AI-automated outreach. They look similar from the outside. They produce very different results.
AI-automated outreach is the volume play: build a list, run it through a template, fire a thousand connection requests, let the numbers work. Some tools and agencies sell this as a service. In a market the size of New Zealand, it’s one of the fastest ways to damage your professional reputation. LinkedIn actively throttles high-volume activity now, networks are small enough that your targets notice patterns, and the messages themselves almost always read like what they are.
AI-assisted outreach is the opposite of volume-first. It’s research-first. The AI does signal intelligence work: scanning a prospect’s recent activity, identifying a company milestone, noting a content post they engaged with, flagging that they’ve recently changed roles. A new head of operations at a Wellington logistics firm who just posted about supply chain efficiency has given you three signals in one move. Writing a connection request that references two of those signals naturally takes thirty seconds with the right tooling. Doing that research manually for forty people a week takes hours.
The output is a message that reads like it was written by someone who did their homework, because it was. The LinkedIn outreach work we do with clients starts here: signal intelligence, not volume.
In NZ and Australian B2B markets, across the campaigns we’ve run, the difference between generic and AI-personalised outreach is significant enough to affect whether the approach is worth running at all.
Generic cold connection requests, the kind with a short templated message or no message at all, run at about 22 to 28% acceptance in these markets. That’s consistent with what most practitioners report for cold LinkedIn work globally.
AI-personalised connection requests, with a short note referencing something real and specific about the person’s situation or recent activity, run at 38 to 46% acceptance in the same markets. The delta is real and repeatable across industries.
The more consequential number is response rate on follow-up messages. A generic sequence, accepted connection followed by a templated value proposition, produces responses from roughly 6 to 9% of people who accepted. An AI-personalised sequence, with a value-first follow-up message that continues the thread of the initial note, produces responses from 18 to 24% in the same markets.
Converting those to conversations: a generic campaign generating 100 connections per month produces roughly 6 to 9 conversations at best, many of which are negative. An AI-personalised campaign generating 60 connections per month, because the quality bar means fewer but better targets, produces 11 to 14 conversations, and nearly all of them positive. Fewer connections, better conversations. Small markets reward that trade every time.
Outreach doesn’t exist in isolation. When someone accepts a connection request and looks at your profile, the profile has to answer the questions your message raised. When they click through to your website, the site has to close the loop on who you are, what you do specifically, and whether you’re credible.
The most common reason AI-assisted LinkedIn outreach underperforms relative to what it should deliver is not the messages. It’s the destination. A personalised, well-researched connection request from someone with a vague LinkedIn profile and a website that says “we partner with clients to drive growth” creates cognitive dissonance. The message was specific. The profile is not. The credibility doesn’t stack, and the prospect quietly moves on.
This is why outreach work is almost always preceded by a review of the profile and the site. Website design optimised for clarity, and AIO that makes the site readable and citable by AI search tools, function as a unit with the outreach layer. The connection request is the invitation. The profile and site are the room the person walks into. If the room is unprepared, the invitation was wasted.
One consistent pattern across campaigns: outreach acceptance and response rates are measurably higher when the sender has been actively posting on LinkedIn in the weeks before outreach begins.
The mechanism is straightforward. When someone receives a connection request, they almost always click through to the profile. If the feed shows recent, relevant content, the person has evidence the sender is active, informed, and worth connecting with. The connection request becomes confirmation of something they can already see rather than a cold approach. If the feed is empty or months old, the request reads as purely transactional.
For professional services businesses in NZ and Australia, this is one of the clearest arguments for consistent LinkedIn content before outreach starts. Two or three posts per week for four weeks creates a visible, recent track record. Paired with a social media marketing strategy that treats LinkedIn as a relationship channel rather than a broadcast channel, content and outreach compound each other in a way that neither does independently.
The content doesn’t need to be elaborate. A short observation about something happening in your industry, a client insight with the identifying details removed, a question about how your network is handling a specific challenge. Specificity matters more than polish. In small markets, posts that demonstrate genuine knowledge of local conditions travel better than content repurposed from US or UK publications. A Wellington commercial property manager writing about the current state of CBD leasing will outperform a generic business development post every week.
The sequence structure for NZ and Australian B2B typically runs longer than US playbooks suggest. Three-day follow-ups that some US practitioners recommend feel aggressive in markets where professional networks are small and people talk.
The structure that holds up: a connection request with a personalised note referencing something specific and real. Not a pitch. One to two sentences that acknowledge what prompted the reach-out and give the other person a reason to accept.
A pause of three to five days follows. Long enough for the person to notice that you’ve been active on LinkedIn since connecting.
Then a value-first message. An observation, an insight, a link to something genuinely relevant to what they’re working on. No ask at this stage. This message establishes whether you’re someone worth the next step.
A pause of seven to ten days.
Then a soft ask. Not “I’d love to get on a call about our services.” Something more like a question that invites a conversation about a specific topic where your work is relevant. In small markets, asking for a thirty-minute call from someone you’ve never met is often the moment the sequence falls apart. The ask needs to feel proportionate to the relationship you’ve actually built.
Total elapsed time from connection to soft ask: two to three weeks. Longer than many clients want, but it’s the cadence that produces actual conversations rather than polite declines.
The limiting factor in most LinkedIn outreach programmes is not strategy. The strategy is tractable. It’s consistency.
Sending forty well-personalised connection requests per week, maintaining the follow-up sequences, posting three times weekly, reviewing responses and iterating the approach. Done every week for ninety days without interruption, this compounds in ways that produce a real pipeline. Done intermittently, with three-week gaps when other work picks up, it produces frustration and a LinkedIn inbox full of stalled sequences.
The clients who get the most from AI-assisted outreach are the ones who treat it as an ongoing operation rather than a campaign. A campaign has a start and an end. LinkedIn relationship-building in a small market is a long game. The decision-maker who was the wrong fit at the time of first contact may be the right fit eighteen months later, after a role change or a shift in priorities. Being the person in their network who engaged with their content thoughtfully across that period is worth something when that moment arrives.
The practical first step for any NZ or Australian B2B business taking this seriously: audit the profile and the website before touching the outreach itself. If the destination isn’t ready, the investment in personalised outreach is partially wasted. Get the room ready before sending the invitations, because in small markets, first impressions compound in both directions.
From automated guest comms to multilingual booking content, here's where the tools are earning their place.
The tools getting traction in real estate and property management look nothing like the early hype.
After a year of hybrid AI video production for NZ clients, here's what works, what doesn't, and where the economics change.