Priority structure in web content creation projects

Most web content projects include both structural and editorial aspects: for example, information needs to be structured to fit the new responsive design, and the current text needs to be updated to meet the basic principles of messaging and corporate identity.

I am often asked in what order it is better to do the work: first create a structure, and then prepare the text, or vice versa? I have never had an unambiguous opinion on this matter, because, in my opinion, this looks like a chicken or egg problem. If work on a project begins with structure, I create content models with substandard information. If, on the contrary, we start with texts, the writers do not understand how much text we need to fill the models, since the models do not yet exist. It seems that both of these areas have the same problems, so I have no reason to give preference to one or the other.

(Please note that I’m not talking about starting without the editorial basis of the project: understanding business goals, creating a message architecture and understanding what the work should lead to is the basis of any project. I’m talking about processing low-quality content - about editing and creating a new text based on these fundamental principles).

First, structure the content, and then rewrite it.

Recently, I completed the second stage of the project, which we organized, concentrating primarily on the structure, and I had mountains of reasons to choose this approach. I believe that an approach that focuses on structure makes sense in most of my projects, and here's why.

Content models are based on what the content is intended for, not what it reports.

In particular, the text present in this project was terrible. Jargon, cliche and almost complete uselessness. How could I build a model of useful content from bad content?

As I worked, I realized that the quality of the text - even if it is terrible - actually does not affect the models. I am not building the model directly from the words contained in the content; instead, I am building it on the basis of the purpose that it should serve. I don’t really care if the description of the restaurant looks like a teenage poem (I apologize to teenagers and poets): this is a description of the restaurant, and we need a short advertising version and a long full version. The description of the premises for banquets should contain photographs taken relatively recently in good lighting, and in the captions to them should be used the appropriate "voice" of the brand, describing why these rooms can be used. In fact, I don’t need to see quality photographs and worthy descriptions in order to create space for them in models.

Development of the influence of structural decisions and the direction of design

An integrated content model will help to fill content with all types of site solutions, from CMS selection to data formatting. Developers can make more appropriate decisions on architecture when they have an idea of ​​what types of interactions exist between types of content, and designers can organize a library of templates that will correspond to the degree of partitioning of the content model. The sooner structuring is completed, the easier it will be to build an integrated design and development plans.

Trying to squeeze substandard content into good models is an extremely compelling reason for editing.

When working on projects focuses on the structural aspects of the work, we want to re-combine content for different channels or create a reasonable responsive interface using structured fields - people often try to convince themselves that the current content is enough to do the work. “Of course, it could be improved, but now we are not up to it.”

I did not see a more effective argument in favor of the importance of doing editorial work than taking an existing copy and seeing how crooked it fills the model, which we have already approved as relevant to our business goals.

The model I recently created included a type of content for posting ad fragments about the company's excellent customer service. When we started moving existing content into a new model, the only text we found for this purpose was limited to the phrases “free ice water” and “polite workers.” We have already stipulated that the main task of the new site is to talk about the brand’s experience, and the realization that the current content absolutely does not fit into this task was a blow below the belt. Therefore, it was necessary to allocate funds for editing.

Content models are easy to duplicate

Waterfall development has lost its position as the best option for content development, in contrast to the development of design and code, so editing text often leads to changes in content models. I can divide one large field into two small ones, or writers can find a place where I missed an important part of the content. Optimization of models is considered an essential part of the process.

In projects in which text editing was performed primarily, I often come across the fact that well-written text simply does not have a place in the model. In the process of structuring information, we can combine two pages into one, or use the same description in three places, and thus the efforts of editors who were expended in changing such a text are simply discarded before they even see the light of day. This is discouraging, and can lead to content creators deciding that I don’t value their time or their work at all.