Media Strategy · Practical analysis

How to choose advertising channels: reading Kantar Media Reactions 2026

A platform’s user count does not tell you whether it belongs in your campaign. Channel selection also depends on the audience, their receptiveness to advertising, the decision you want to influence and what a local test can establish.

A phone, headphones, magazine and screen on a media planning table
Key takeaway

Use advertising preference reports to form hypotheses, then test them with your own data. Stated preference is not a guarantee of sales.

What does Kantar’s report add?

Kantar Media Reactions 2026 examines how consumers and marketers perceive advertising environments and channels. This article uses its public overview; the full rankings and data are available in separate paid packages that we have not reviewed.

The public page highlights marketers’ enthusiasm for investing in AI assistants alongside lower consumer trust in their recommendations. That raises a question about audience readiness, rather than automatically making the case for spending. We do not reproduce unverified country or platform rankings. [1]

Separate reach, receptiveness and results

Reach describes whom a channel can expose to a message. Advertising receptiveness describes how willing people are to engage with advertising in that environment. Business outcomes describe what happens afterwards. These are related, but they are not interchangeable.

A channel with inexpensive, broad reach may attract people who do not need your service. A smaller specialist channel may capture existing demand. Assign each channel a role: introduce the problem, explain an offer, capture intent or reconnect with interested visitors.

Write the channel brief before choosing a platform

QuestionWhat to define
Whom are we trying to reach?Market, language, problem and buying stage.
What is the channel’s role?A measurable objective that distinguishes awareness from sales.
What content does it need?Suitable video, explanation and destination page.
What counts as quality?A shared definition of a qualified enquiry or completed sale.
When will we decide?A spending limit and a learning period.

A global report does not replace checking whether your offer, content, page and follow-up are ready for the audience you want to reach.

Design a test that answers a local question

Consider a hypothetical academy serving Arabic-speaking professionals in Turkey. A search channel can capture people already looking for training; an explanatory channel can introduce its value. Comparing clicks alone would ignore their different jobs.

Keep the offer consistent, adapt content to the channel and use the same qualification and follow-up standards. Add tracking parameters and connect them to enquiry records. Our GA4 measurement guide explains the measurement side.

A small budget spread across too many platforms may produce too little evidence anywhere. Define what would make you continue, change or stop each test before it begins.

Do not let the lowest lead cost make the decision

Review cost per qualified lead and progression towards a sale. With a long sales cycle, early indicators are provisional evidence, not a final return-on-investment calculation.

Advertising platforms may each claim the same conversion. Reconcile results with your customer records rather than adding platform totals together. Produce a one-page review for each channel: its role, what you learned, spending, lead quality and the next proposed change.

Frequently asked questions

Does the report identify the best platform for every business?

No. Interpret it in the context of your market, offer and objective, then validate the hypothesis with local data.

Should every channel receive the same budget?

Not necessarily. Budget should reflect its role, the hypothesis and the amount of data needed to make a useful decision.

Sources and editorial review

Original sources used in this article. Last editorial review: 1 October 2026.

  1. Kantar Media Reactions 2026 — public overview2026 edition; public overview reviewed 1 October
Back to the top