From Psychology To Marketing
June 22, 2016
From Psychology To Marketing
Score 8 out of 10
Vetted Review
Verified User
Modules Used
- IBM SPSS Statistics
Overall Satisfaction with IBM SPSS
It was recently discontinued as the license available expired following the departure of a BI manager. It was mainly used by myself as an attempt to provide statistical rigour using inferential statistics to problems which were usually only approached using descriptive statistics. It was not widely used across the organisation and only in Marketing/Bi as far as I am aware.
- It provides incredibly accurate and robust results.
- The print-out is incredibly detailed and allows you easy understanding of the post-hoc results and can be saved and easily re-examined at a later date should someone query the findings.
- It is good for saving time by being able to add multiple descriptive statics alongside your inferential etc., at the touch of a button.
- Some of the lexicon can be confusing, especially for those who are not out-and-out statisticians. Some people may use the term MANOVA or ANOVA instead or 2-way multivariate etc. There sometimes has to be a lot of Googling to quickly find the test that's right for that situation.
- The output, whilst functional, isn't very attractive or necessarily easy to output to other BI software packages which are more aesthetically pleasing or easy to digest for non-statistically minded audiences.
- The input is not always as easy as can be in organisations which are not set-up for One-Subject-One-Row data input and this can involve lots of manipulation. Perhaps more trans positioning tools to take data from Excel, Access or if they are present make them more obvios.
- This is difficult to quantify, but the results helped shift the approach from descriptive-based statistics reporting to inferential. This was particularly useful in judging the impact of CRO/UX changes made on the website and their likely benefits.
- It's been used as a portfolio of arguments for where to place channel investment based on predictive regression models and this was seen as one of the most persuasive parts of an argument which then went on to prove effective.
- Analysis carried out using SPSS in conjunction with a CRO package has led to an estimated 12% increase in online conversion.
- R-project and KNIME
I used SPSS throughout my degree and it was the software package I was most comfortable with. Beyond this, it doesn't require any C-style coding ability like R needs and therefore is very user-friendly to the layman statistician. Beyond this, it contains every major statistical manipulation you are likely to require within a Marketing/BI environment and with the simple data input style and detailed output, you know you have a very high likelihood of getting results correct 100% of the time