Almost every influencer proposal a New Zealand brand receives is organised by tier. Nano, micro, mid, macro, mega, with a rate attached to each and a chart showing reach going up as you move to the right.
The chart is accurate. It is also the wrong chart, because reach is an input and nobody buys reach for its own sake.
What the published research actually found
Two studies in the Journal of Marketing are worth more than the entire category's supply of case studies, because they used campaign data rather than self-reported results.
Engagement falls as audience grows
Wies, Bleier and Edeling analysed 802 campaigns and published the result in the Journal of Marketing in 2023. Engagement with sponsored content follows an inverted U against follower count. Reach rises with audience size, but the likelihood that any individual engages falls.
The mechanism is not mysterious. A post from someone with 900 followers reaches people who know them. A post from someone with 900,000 reaches people who follow a media property. The second is broadcasting, and audiences respond to broadcasting the way they always have.
The economics point the same way
Beichert, Bayerl, Goldenberg and Lanz, publishing in the Journal of Marketing in 2024, compared campaign economics across the tiers. Nano-influencers generated EUR 0.265 of revenue per follower against EUR 0.049 for macro-influencers. Return on influencer spend came out at 17.85 against 4.67.
Those are averages from someone else's market, and they are not a promise. What they establish is direction, and the direction is the opposite of the way most rate cards are built.
Why the market keeps selling the other end
Three reasons, none of them sinister.
Reach is easy to sell. A number in a proposal that goes up is easier to approve than an argument about audience composition.
Nano is harder to run. Fifty small campaigns need a system: recruitment, matching, verification, moderation, rights, payment. One large campaign needs a contract and a courier. Agencies are rational about this.
Nano is harder to case study. There is no famous name to put on the slide at the end.
What nano is not
It is not a cheaper version of the same thing. Taking a macro brief, dividing the budget by forty and sending the same caption to forty small accounts destroys the exact mechanism that makes small audiences work, which is that the audience knows the person and can tell whether they mean it.
It is also not about follower count as a target. We do not set a minimum, because the number is a proxy for the thing that actually matters, and the proxy is weak. What matters is whether the audience is made up of people who would ask this person for a recommendation anyway.
How we run the far end of that curve
In a nano campaign here, participants are matched from the community on category purchase behaviour and lifestyle data rather than on audience size. They buy the product themselves. Proof of purchase is matched to the customer and to the post. The advertising label goes at first interaction, because the brand paid for it.
Then organic performance decides which posts get media behind them. That is the step that solves the reach problem the tier chart was trying to solve. Small audiences supply the credibility; paid amplification supplies the scale, and it does it after the content has already been tested on a real audience rather than before.
The New Zealand read
In a New Zealand campaign for Kiwicare Weed Weapon in December 2025, five Canterbury gardeners took part. No follower requirement, no rate card tier. Their content ran as paid media alongside the brand's own advertising in the same campaign: 2.78% click-through against 1.07%, 4,937 clicks at NZ$0.20 each, and a reported 88% of campaign reach in the 25 to 54 age band.
Source: Kiwicare Campaign Report, January 2026.
Five people does not make a study. One campaign, in one category, in one market. It is also the only New Zealand comparison of its kind we can put our name to, and it points the same way the published research does.
The question to ask a proposal
Not how many people will see this. Ask how many of the people who see it know the person posting it, and what happens to the content that performs.
A proposal that cannot answer the second question is selling you impressions, whatever tier it is priced at.