Predicting Your Content’s Influence on the User

Users will always have some reaction to a webpage or application. They might move on to another page, read part of the content and ignore the rest, change their lifestyle (diet, exercise routine or sleeping habits) or make a purchase. Despite extensive research into content components (including their relation to specific target audiences), there is as of yet no model to explain why a user will react to a given page in a certain way.

Predicting the user’s reaction is particularly important to providers of health, medicine and wellbeing content, as well as e-commerce websites.

A study conducted by Macabbi Healthcare Services together with Dr. Rita Mano revealed that 27.7% of the study’s participants found information online that encouraged them to change their medical routine; 19.7% found information that convinced them to start exercising; 7.6% were persuaded not to take medication recommended by their doctor and 4.4% decided to switch doctors following new information they read online. The study found that users’ reactions were dependant on their condition.

If there was a formula to calculate the influence of all content components on a specific target audience, this could help architects and writers identify any weak links in their page content and thus drastically improve user reactions by altering them.

Using numerical analysis to create such a formula has been common practice in a variety of fields for many years now. My own thesis used this technique to predict the effects of different amino acid levels in food on the animal that ingested it.

A Formula to Predict a Page’s Influence on a Specific User

Here’s an example of how a formula like this might work.

  1. Let’s take a specific content variable, say the “Phrasing” of a page. Its value is determined by calculating the users’ subjective evaluation of it together with objective scientific criteria. Then this value will be multiplied by its relative significance (comparable to the significance of other content variables).
  2. Now let’s assume our target audience is hypochondriacs. The “Phrasing” variable carries a predetermined reference value for hypochondriacs, so we will multiply that value by the by its significance (i.e., the extent to which “Phrasing” influences hypochondriacs).
  3. Let’s say our hypochondriacs have a high school level of education. High school education has its own predetermined reference value, so we will multiply it by its significance (its influence where “Phrasing” is concerned).
  4. We will continue with this procedure for all the variables found to be significant in relation to “Phrasing”.
  5. Finally we will multiply the values we’ve just calculated and are relevant to “Phrasing”, by the “Phrasing” value we determined in step 1.
  6. This process (1-5) will be used for each of our content variables. Then we will aggregate the resulting values to produce a numerical value which represents our content’s influence on specific hypochondriacs. I predict that specific influence values will be linked to certain degrees of influence, for example that exposure to the material will lead users to seek professional consultation. 

It’s possible to operate this model’s equation on other target audiences and examine the results.

 

Here’s how it works (by “Content Characteristics” I mean whether the content is textual, textual with visual aids, mostly visual, etc.)

 

Here's how it works

Here’s how it works

 

The Formula’s Commercial Potential and Why it Hasn’t Been Developed

Once the formula has been developed and validated (following extensive research), it will allow us to use a simple sensitivity analysis to determine the weak link(s) in our content and predict how improving this component will affect a specific user’s reaction to the page. This obviously has commercial benefits.

The main obstacle in the development of this formula isn’t technological, but conceptual and budgetary. Unlike the fields of biotechnological or medical research, where every dean known that important research requires a budget, investment in information technology research seems to be sorely lacking in Israel. The export of Israeli knowledge was for many years a commercial success, but only because of the vision and efforts devoted to it.

 

 

Member of