What materials should a business prepare before starting an AI marketing project?

Start with objectives and audience information

Before introducing tools, explain whom the project is intended to help and what tasks it should help them complete. Identify the planned publishing channels, responsible people and criteria for success. The objective should go beyond simply “producing more content” to define the specific problem and how the content should support decisions.

Drawing on publicly discussed CES topics concerning trust and transparency in intelligent content, this article offers independent advice on project preparation. Businesses should determine their tools and business objectives based on their actual circumstances.

Organize facts and versions

Prepare the company overview, product names, model numbers, service scope, applicable conditions, limitations and current contact information. Label each item with its source, reviewer, version and last updated date. Keep historical materials and unconfirmed plans separate from currently valid facts.

For customer case studies, data, awards and credentials, confirm whom or what they concern, the permitted scope of use and the relevant authorization. Do not include unsupported rankings, claims about customer results or performance promises in the asset package.

Create a list of missing information: whom to ask, when confirmation is expected and which statements must not be published externally before confirmation.

Confirm asset rights and tool boundaries

Record the sources of logos, fonts, images, videos, customer testimonials and research materials, along with the channels and periods for which their use is permitted. Keep assets with unclear permissions out of generation tools, and do not assume that purchased assets can be used in every format.

Confirm which inputs the tools will access, how data will be processed, what permissions team members will have and whether the tools are suitable for customer or internal materials. Review them against the business’s actual privacy, confidentiality and supplier management requirements.

Create brand voice guidelines and examples of output formats, but remember that these materials cannot replace fact-checking or asset permissions.

Assign human responsibility and establish a publishing process

Define who organizes the materials, who generates the first draft, who checks business facts, who confirms asset usage, who approves publication and how errors will be corrected. High-risk content requires appropriate professional review; “AI-generated” must not be treated as an approval status.

Retain versions of input materials, draft revisions and approval records. Clearly label unreviewed drafts to prevent other teams from using them by mistake. When reusing content across channels, check that its context, conditions and links still apply.

Test the process first with a small number of real tasks, including cases involving insufficient information, conflicting versions and assets for which permission cannot be obtained. Confirm that the team knows how to pause work or escalate issues.

Let the quality of preparation determine whether to proceed

Before starting, ask the relevant people to spot-check whether a product statement can be traced to its source, an image’s usage permission can be confirmed and a draft’s approver can be identified. If any critical issue remains unresolved, complete the materials or narrow the project scope first.

Qingmei Future can assist with content planning and brand workflow design; the business remains responsible for business facts, data permissions, asset rights and final publication.

For example, having a designer create a product image does not mean it can be used on every advertising platform. Asset licenses, customer logos and cropping requirements for each channel still need to be checked. Assign a storage location and contact person for each type of material so the team can still find the authoritative version when employees leave or suppliers change.

Prepare one package of materials cleared for public use and a separate checklist for internal review only, with clear differences in access permissions. Have the responsible people verify any unclear data processing terms from tool providers before proceeding.

Before the pilot, check the channels where generated content will subsequently be used, ensuring that draft labels and approval status are preserved during export.

When starting the project, appoint a materials coordinator to resolve version conflicts, authorization questions and items awaiting confirmation, reducing the need for team members to search individually for old files.

Also define which members can view project materials to prevent case studies with unclear permissions from being copied freely between departments.

After the pilot begins, record the information gaps and review times encountered in practice, and promptly complete the source materials instead of expanding the project scope indefinitely.

Discussion Background and Scope

CES · AI or Human? Trust, Transparency, and Co-Creation in the Age of Intelligent Content · 2026-01-07. Official page date: 2026-01-07. The source discusses authenticity, transparency and accountability in intelligent content. It is background reading, not evidence for each practical recommendation in this article. The recommendations are Qingmei Future’s independent analysis.

Related Services and Content

Explore AI marketing services　Explore brand strategy services　Discuss your content workflow