Conty expands campaign scopes: different deliverables call for different evaluations

Being in the same campaign does not make every assignment the same

A brand might commission a detailed demonstration, answers to questions and a post introducing a product for the first time. They all contribute to the same commercial objective, but do not necessarily have the same assignment. Adding their results together can help describe the operation. Ranking creators by a single column can hide what each was asked to do.

The problem starts before the report, when the proposal describes only formats and quantities. “One video” does not say whether the creator must research a question, test an assembly, explain a difference or spark curiosity. Without that definition, the team risks evaluating an assignment that was never agreed upon. Content commissioned to clarify something should not be treated as a failure simply because it did not behave like a piece designed for reach.

In Renoir's analysis, evaluation becomes clearer when it separates three questions: did the deliverable fulfill the agreement, does the audience response provide useful signals about its function, and is there a basis for comparing it with another piece? A positive answer to the first does not guarantee the second. The third depends on conditions that may be absent even when plenty of numbers are available.

Two creators and a desk lamp

Consider a hypothetical campaign for a lamp with adjustable positioning and different lighting controls. One creator is asked to show how she organizes a reading space. Another is commissioned to answer recurring questions about the controls and demonstrate the sequence of use. The first piece introduces a recognizable situation; the second needs to make an operation understandable.

For the setting demonstration, the team wants to check whether the audience understands when the product fits into a routine. For the explanatory content, it wants to know whether the question was answered without leaving out an important step. These functions should guide preparation and interpretation of the result. They do not mean that one creator is forbidden from reaching many people or that the other cannot teach; they identify the main responsibility of each deliverable.

After publication, the first piece receives more views. The second receives fewer, but some comments describe questions being resolved while others indicate confusion about a particular control. Declaring the first creator the winner on audience size does not establish whether the second creator's explanation worked. Declaring the second the winner on comment quality would also be premature: commenters may represent only a portion of viewers.

The team can return to the content and locate the reported problem. If the video shows a hand covering the button just when the action should be visible, there is a concrete hypothesis of a flaw in the explanation. If the operation is clear and the question concerns another lamp model, the response calls for clarifying the difference, not automatically remaking the demonstration. The comment is evidence to investigate, not a final performance score.

The report needs to show the conditions behind the numbers

Before comparing, record relevant differences between deliverables: format, duration, publication timing, audience size and composition when known, paid media support and the piece's place in the campaign. These factors do not explain everything, but they help avoid easy conclusions. A boosted post and one distributed only organically do not provide the same observation conditions.

This does not require inventing a formula that corrects every difference. Dividing results by follower counts, for example, does not make two audiences equivalent or remove the effect of distribution. Sometimes the appropriate conclusion is that the pieces performed different tasks and do not support a direct ranking. It is still possible to decide what to keep, fix or investigate without pretending the comparison answers more than it does.

In the example, the report can present the completed deliverable, observed signals and proposed decision separately. For the setting piece, record available reach and questions about use. For the explanation, record the passages that prompted questions and the signs of understanding found. If visit or purchase data exists, state its source and what it allows the team to attribute. Comments, views and sales should not be added together as though they represented the same action.

The next investment needs a specific question

Compensation does not become comparable simply because both deliverables are videos, either. Preparing a demonstration may involve different work from recording an everyday situation. The amount per post describes a cost, but does not by itself explain the effort commissioned or the result achieved. The decision must consider the agreed scope and what the campaign still needs to resolve, without turning the example into a price list.

If the team wants to know which approach explains the controls better, it can plan another comparison with the same assignment and more similar conditions. If it wants to evaluate whether the explanation complements the setting presentation, the question is different: does the combination answer the audience's questions better? Changing the question after looking at the results makes it easy to choose a winner and difficult to learn what to repeat.

There is also a boundary of responsibility. A creator may deliver a correct explanation while the linked destination provides confusing product information. The team needs to investigate that journey before attributing the entire result to the piece. Likewise, fulfilling the scope does not rule out editorial revision when the material does not help the audience. The analysis should locate the problem and who can fix it, rather than search for a single person to blame.

Campaigns with different scopes allow tasks to be distributed according to what each contribution needs to offer. The report should preserve that logic. The most useful evaluation makes clear what work was completed, what was observed and which decision those signals support. When there is no basis for a direct comparison, acknowledging the limit helps improve the next commission without reducing distinct contributions to a contest in the same column.