The brand influence assessment method for thought and discussion

Initially, poly-cost-effective in all types of media, a large number of ads, and quickly spread a cost-effective brand image, which in non-Taobao users to cultivate their own brand awareness, Taobao users also consolidate the poly Cost-effective understanding of the brand, so the value-added together to enhance the brand played a very positive role.

In advertising before, using research team Q3 had a cost-effective brand influence were investigated in the study focused on six main indicators:

- Familiarity: Examining whether or not the user knows and understands the extent of the user to the brand, which reflects the market's performance and status

- Popularity: To examine whether the popular / popular brand, which reflects the market's performance and status

- Relevance: Examine the brand and whether the product is suitable for the target population, which is the basis of brand equity

- Uniqueness: The brand is different from other brands, reflecting the brand's ability to surpass its competitors and the brand's future

- High Quality: User-perceived quality is the most basic element of brand long-term survival

- Degree of trust: Whether the user trusts or not reflects the brand's ability to bring the interests and rights of users

These six indicators in the market research industry, widely used in brand research, and each indicator with a sentence expression, the formation of scales, the cost of user input is not. This transplanted to the Internet innovative product research, the basic can achieve the desired research purposes. This article does not focus on cost-effective brand influence in-depth analysis, but focused on the analysis of the method.

Dimension reduction

In this case, we use the six indicators to evaluate the impact of the poly-cost brand, but we can find a strong correlation between the six indicators (the correlation coefficients exceed 0.5 and are very significantly positive). In the comprehensive evaluation, Consider simplifying the indicator.

In general, the use of principal component analysis or factor analysis. In most cases, the principal component analysis obtained by principal component analysis is not easy to explain, and factor rotation can reduce the common ambiguity in the principal components, which can make the structure of factor loading simpler and thus more conducive to factor interpretation. In addition, the commonly used software can directly output the value of the factor score, and the value of the principal component needs to be manually calculated again.

First of all, the six indicators of factor analysis, the resulting KMO value of 0.891, Bartlett test the value of 2747210.409 (df = 15) reached significant, very suitable for factor analysis. Finally, two common factors were extracted, the cumulative variance contribution rate was 84.6% (see the table below), the interpretation is very strong; the ratio between the original variable matrix and the reconstruction matrix> 0.05 is 26% Co-effect is better.

Through the factor loading matrix, we can find that the first common factor focuses more on the psychological cognition after polygration, while the second common factor more reflects the perception of the facts and phenomena. Therefore, the first common factor is named as the deep characteristic and the second The common factor is named the surface feature.

In addition, the correlation indexes have relatively large loads on the two common factors. Generally, when the validity of construction is considered, the indexes with larger loads on different common factors will be deleted. However, this example does not seek to construct validity , But instead want to retain the indicators in the index system as far as possible to achieve the purpose of dimensionality reduction for further analysis. Therefore, the relevant indicators have not been deleted.

Calculate the comprehensive score of brand influence

With two common factor scores, still can not do a comprehensive assessment. In order to be more simplified, the final use of a single data to reflect the brand's comprehensive influence needs further calculations.

The comprehensive score of factor analysis needs to use the eigenvalue and factor scores of each factor. The calculation equation in this example is as follows:

The composite score for a single sample = (feature root 1) / (feature root 1 + feature root 2) * common factor 1 score + (feature root 2) / (feature root 1 + feature root 2) * common factor 2 score

Because the factor score is the standardized score, the calculated comprehensive score is also the standardized score, which needs to be converted into a number of 0-100 so as to intuitively reflect the comprehensive influence of the brand. Need to be treated as follows:

Score of single sample brand comprehensive influence = (composite score of single sample - minimum of composite scores of all samples) / (maximum of all composite scores - minimum) * 100

In this way, each sample will be able to get a comprehensive cosmopolitan score for the poly-cost-effective and conduct a comparative analysis among different groups.

Comparison of brand influence differences between groups

First of all, make a difference test between male and female groups. T test by independent samples shows that there is a significant difference between men and women on the influence of the cost-effective brand, and men's higher comprehensive evaluation of the cost-effective;

Second, different groups of cities at different levels of testing. The chi-square test (comparison between two pairs) shows that users in different cities have significant differences in influential brands. Users in third-tier cities have the highest comprehensive evaluation on the influence of cost-effective brands, while those in first-tier cities have low comprehensive evaluation.

After the poly business adjustment and brand advertising dissemination, I believe the brand influence score will have a more substantial increase, pending follow-up study verification.

summary

Brand influence assessment method, the basic operation process is as follows:

1, to determine the assessment of the sub-indicators;

2, through factor analysis (or principal component analysis), to reduce the dimension of sub-indicators, and calculate the common factor score;

3, calculate the composite score of the common factor, and translate into visual scores of 0-100;

4, compared to different groups of brand influence comprehensive score.

When comparing the impact of different brands, or different stages of the same brand, you need to put the data together, the above calculation process. Only in this way can we keep the same level of comparison.