Brands asking how to run an instagram influencer campaign almost always brief toward engagement as though it were one outcome, and a 2026 analysis of 15 million Instagram posts drawn from 417,130 active pages found that engagement is not one behavior at all. It is three, and each belongs to a different format. Reels lead on comment rate at 0.06% against 0.04% for carousels and 0.03% for images. Reels lead on share rate at 0.10%, with carousels at 0.08% and images at 0.07%. Carousels lead on save rate at 0.05%, ahead of Reels at 0.04% and images at 0.02%. The median Reel earns 5 shares where the median carousel earns 3 and the median image earns 1. A brief that asks a creator for engagement has therefore specified nothing. It has delegated the campaign’s objective to whichever format that creator happens to post out of habit.
Table of Contents
- Why the Split Between Comments, Shares, and Saves Reframes the Brief
- What Enterprise Brands Should Expect From an Instagram Influencer Campaign Partner
- Program Delivery Across Instagram Influencer Campaigns
- How to Evaluate an Instagram Influencer Campaign Agency
- The HireInfluence Model for Instagram Influencer Campaigns
Why the Split Between Comments, Shares, and Saves Reframes the Brief
The single engagement number was useful when it was the only number available. It is a ratio, it is comparable across accounts, and it produces something a reporting deck can hold. What it cannot do is distinguish between an audience that is arguing with the content, an audience that is forwarding it to someone else, and an audience that is filing it away for later. Those are three unrelated dispositions toward a brand. Collapsing them into one figure does not summarize the campaign. It deletes the campaign’s result.
Shares are the only one of the three that does structural work a brand cannot buy another way. A comment happens inside the existing audience. A save happens inside the existing audience. A share is the content leaving the creator’s follower graph and arriving in front of someone who never opted into it, carried by a person who chose to put their own name behind the delivery. That is earned distribution in the literal sense, and the format data says it is not evenly available. The 0.10 percent share rate on Reels against 0.07 percent on images is not a small edge on a vanity metric. It is the difference between a campaign that compounds beyond its own reach and one that does not.
Saves point at something else entirely, and the carousel’s lead there is the most useful finding in the study for a brand with a considered purchase. A save is a declaration that the content will be needed again, which only happens when the content contains something worth returning to. The carousel wins saves at 0.05 percent because the format holds more information than a viewer can absorb in one pass, and that is a statement about the content’s function rather than its aesthetics. A brand whose product requires a decision rather than an impulse has a natural format, and it is not the one most campaign briefs default to. The default is the Reel, because the Reel is what the category talks about, and the category talks about it because it is the format that produces the most visible motion. Visibility and usefulness are unrelated properties.
The comment finding is the one that most often gets misread. Reels lead comments at 0.06 percent, and the intuitive explanation is that video is simply more engaging. The more precise reading is that a Reel creates the impression of a person addressing the viewer directly, and direct address invites reply. That means comment volume on a Reel is partly a property of the creator’s delivery, not of the brand’s message. A campaign that is measured on comments is, to a meaningful degree, measuring the creator’s on-camera manner. That may be exactly what a brand wants. It should at least be a decision.
None of this makes one format better. It makes the format decision identical to the objective decision, and that is the sequencing error the data exposes. In most programs the creator is selected first, on audience or aesthetic grounds, and the format arrives as a byproduct of that creator’s habits. The campaign then reports whichever behavior the habit happened to produce and calls it performance. Running the sequence in the correct order means naming the behavior the brand needs, deriving the format from it, and only then selecting creators who are credible in that format. Legibility of the goal has to precede the casting.
What Enterprise Brands Should Expect From an Instagram Influencer Campaign Partner
Program strategy and design. The agency has to convert a business objective into a named engagement behavior before it approaches a single creator, because reach, reply, and retention are separate outcomes with separate formats attached. This is where dedicated campaign services either produce a campaign with a thesis or produce a posting schedule with a budget.
Creator sourcing and verification. The agency has to evaluate creators against the format the objective requires rather than against a general engagement score, since a creator who is excellent at carousels and a creator who is excellent at Reels are not interchangeable assets. Verification includes confirming that a creator’s format history matches what the brief will ask them to make.
Platform and commerce integration. The agency has to know how each format behaves once a shopping surface is attached to it, because the tolerances that govern a carousel built to be saved are different from those governing a Reel built to be forwarded. Integration decisions belong at the brief stage rather than after the asset exists.
Creative direction and content production. The agency has to direct toward the behavior the format actually produces, which means a save-oriented carousel needs genuine reference value and a share-oriented Reel needs a reason for a viewer to attach their name to it. The distinction between contributed and commissioned material is set out in the UGC overview, and it governs how much of the creative a brand can direct without collapsing the effect.
Audience and segment-specific execution. The agency has to recognize that the same format serves different segments differently, and that a campaign built to be forwarded and a campaign built to be filed are aimed at people in different stages of a decision. Segment logic determines format logic rather than following it.
Cross-platform orchestration. The agency has to sequence Instagram against the channels around it rather than treating each as a separate program, and adjacent reading such as this TikTok influencer marketing resource clarifies why an asset that performs on one surface can fail on another for reasons that have nothing to do with its quality. Orchestration is what keeps a format decision from being made twice.
Paid amplification. The agency has to select amplification candidates on the behavior a post earned rather than on its total interaction count, because a post that earned shares and a post that earned likes are telling a media buyer different things. A specialist specialties and services capability matters at the point where organic signal has to survive translation into a paid unit.
Attribution and measurement. The agency has to report comments, shares, and saves as distinct results rather than summing them into a single rate, since the sum is precisely the figure that hides which of the three the campaign achieved. An analytics capability that cannot separate them cannot tell a brand whether the program worked.
Program Delivery Across Instagram Influencer Campaigns
Program delivery is where format discipline either holds across a roster or dissolves into whatever each creator prefers. The Grammarly creator program ran 133 creators to 214M impressions, 33.1M views, and $15M in earned media value, a scale at which format drift stops being a rounding error and becomes the difference between two entirely different campaigns running under one name. The Southwest Airlines #SouthwestSaysAloha program produced 56M impressions and 3M engagements. The MTV #MyMTVStyle campaign delivered 16.1M impressions and 216,600 engagements at $0.01 CPV and $1.50 CPM, figures that only mean something once the underlying behavior is named. The Oreo and McDonald’s #OREOShamROCKout activation reached 1.7M impressions at $0.06 cost per engagement. The Ricola case study shows the full chain from behavior to outcome: 26M impressions and 20.5M reach at a 13.17% engagement rate across 18 influencers, resolving into 62,500 MikMak retail clicks, which is an engagement figure that survived translation into an action. Additional programs across formats and categories are collected in the work portfolio.
How to Evaluate an Instagram Influencer Campaign Agency
First, ask which engagement behavior the campaign is for. The agency should answer with one of comments, shares, or saves rather than with the word engagement, and should explain what follows from the choice.
Second, ask how format is decided. The agency should derive format from the objective and then cast against it, and should be able to describe a program where a strong creator was passed over on format grounds.
Third, ask what a share is worth to this brand. The agency should treat shares as earned distribution beyond the follower graph, and should be able to say whether that is the outcome this particular program needs.
Fourth, ask how the three behaviors are reported. The agency should present them separately, and should be able to explain why a blended engagement rate would obscure the result.
Fifth, ask what the program costs and how the number decomposes. The agency should account for creator fees, production, usage rights, and management as separate lines, and the logic underneath those components is laid out in this cost of influencer marketing guide.
The HireInfluence Model for Instagram Influencer Campaigns
HireInfluence has run as a full-service enterprise influencer marketing agency since 2011, and its about section records a team of 25 or more spread across 10 or more states, with offices in Houston, The Woodlands, Austin, Los Angeles, and New York. The firm operates at a six-figure engagement floor, which is what pays for measurement infrastructure capable of separating three engagement behaviors instead of averaging them into one reportable number. Programs have run for Microsoft, Meta, Ricola, McDonald’s, Southwest Airlines, and Oreo. The firm was named Marketing Agency of the Year at the 2024 MUSE Creative Awards and Digital Marketing Agency of the Year at the 2026 U.S. Agency Awards, and it has been a TikTok Shop Lite Program partner since July 2024. Brands scoping an Instagram program can reach the team through the contact page.
Founder and CEO Jason Pampell spent years managing content rights, licensing, and strategic media partnerships at Forbes and Billboard before founding the firm in 2011, and the discipline he practiced there is the direct ancestor of the format problem this research describes. A licensing desk never asks whether a property is good. It asks what the property is for, because a photograph cleared for a cover and the same photograph cleared for a poster are different assets governed by different terms, and confusing them is how a rights holder gives away value without noticing. Matching talent to property means knowing the use before the negotiation opens. A brand that casts a creator before naming the behavior it needs has made the same error in the other direction.
The benchmark research makes the final case on its own terms. When a platform has three engagement behaviors and a brief has one word for all of them, the campaign is not being measured against its objective, because it was never given one.
