The follower count is the last metric in influencer marketing that has survived purely on institutional habit. Everyone in the industry knows it is a poor predictor of outcomes. It still anchors nearly every deal, every rate card, and every conversation with a CFO.
Nano influencers, meaning creators with roughly one to ten thousand followers, are where that assumption breaks most visibly, and they are worth understanding properly rather than as a budget-conscious compromise.
Engagement rate declines fairly reliably as audience size grows. Nano creators regularly post engagement rates several times higher than mega-influencers, and on some platforms the gap is dramatic enough to be uncomfortable for anyone who has been buying reach.
This is not a quirk of the data. It follows from what happens to a community as it scales.
A creator with four thousand followers can read every comment and reply to most of them. Their audience knows this, and behaves accordingly, because interaction has a reasonable expectation of being seen. Commenting is worth the effort because someone is actually there.
A creator with four million cannot read the comments, and the audience knows that too. The relationship becomes broadcast rather than conversation, and the behaviors that constitute engagement stop making sense for the person performing them. Why comment into a void.
The trust that follows from that intimacy is the actual product being sold. A nano creator recommending a product is structurally closer to a friend recommending it than to an advertisement, and the conversion behavior reflects that. Their audience has not learned to discount their recommendations yet, because they have not been sold to a hundred times.
We should be honest about why most brands do not do this, because the reasons are not stupid.
The reach per creator is genuinely small, and the operational cost per creator is not proportionally smaller.
Briefing, contracting, chasing, reviewing, and paying a creator takes roughly the same effort whether they have four thousand followers or four hundred thousand. If that overhead is four hours per creator, then a fifty-creator program costs two hundred hours of someone's time, which is more than a month of full-time work, before a single piece of content exists.
This is the entire reason the tactic remains underused. It is not that brands do not believe the engagement data. It is that running fifty relationships is a real job, and most teams do not have anyone to do it, so the tactic gets approved, half-executed, and abandoned.
Quality variance is higher. Nano creators are frequently not professionals, which means production quality is inconsistent, deadlines slip, and some percentage of them will simply not deliver at all.
You have to budget for attrition. If you need forty pieces of content, you should be signing fifty-five creators, and treating the non-delivery as a line item rather than a crisis.
The reporting looks bad. A nano program produces a spreadsheet full of unimpressive individual numbers, and an executive scanning it will see fifty rows of four hundred views. The aggregate may be excellent and the presentation is inherently unflattering, which is a political problem rather than a marketing one, and it has killed a great many working programs.
Standardize everything you can. One brief, one contract template, one set of deliverables, one payment process.
The moment you start customizing per creator, the model stops scaling, and the operational cost that was already the binding constraint becomes fatal. Accept that standardization costs you some quality at the margins and buys you the ability to run the program at all.
Use product plus performance, not just cash. Many nano creators are genuinely motivated by product and by the legitimacy of a brand partnership, which is a real asset for them.
A modest fee plus product plus an affiliate rate is frequently more attractive to a nano creator than a flat fee, and it aligns the incentives properly. It also keeps your fixed costs down, which is what makes the volume possible.
Concentrate them in time. Fifty nano creators posting across six months is diffuse and produces nothing you can point at.
Fifty posting in the same two-week window creates the appearance of a movement, and that concentration is where the value comes from. Someone who sees your product mentioned by three creators they follow in one week does not experience that as three ads. They experience it as a thing that is happening.
Track by cohort, not by creator. Individual nano posts will look unimpressive, and evaluating any single one against a macro benchmark will make you cancel something that is working.
The unit of analysis is the program. What did fifty creators cost, in money and in hours, and what did they produce in aggregate, in reach, in content assets, and in tracked conversions. Judge that against what the same budget would have bought from one macro creator, and be honest about the comparison in both directions.
The criteria we actually use to evaluate a creator, in order of weight.
Audience relevance. How precisely their followers overlap with the buyer. This is the single strongest predictor of performance and it is not close.
Engagement quality. Not the rate, the quality. Do the comments contain sentences written by humans who appear to have watched the content, or is it emoji spam and reciprocal engagement pods.
Content quality. Can they actually make something good. A creator with a perfect audience who produces unwatchable video is not useful.
Conversion history, if it exists. This trumps everything else. A creator who has demonstrably sold product for a comparable brand is worth several times one who has not, regardless of any other metric.
Follower count appears nowhere on that list. Its only legitimate use is as a rough constraint on how many creators you need to hit a reach target, which is an arithmetic function rather than a judgment. As a proxy for value it has been wrong for years, and it will keep being wrong for as long as it remains the easiest number to see.