Hands-On Data Science for Marketing: Improve your marketing strategies with machine learning using Python and R

I bought this book having spent the last 5 years working in data analytics and it was good to see something specific to marketing. I enjoyed going through the coding examples (every chapter features code in both Python and R). The code was about 90 percent accurate with a few typos or misprints here and there. I would encourage you to download the answers and try out the examples in one or both programming languages. I certainly see where you could use these examples, learn and either put together something on your laptop or work on your favorite public cloud. You do need at least some understanding of programming and willingness and ability to debug coding mistakes. This was a great tutorial if you have the willingness to roll up your sleeves and write some code.

Easily a book that could have been a five-star contender, unfortunately, there are some shortcomings.Overall, the text is a fine (but basic) introduction to data science through applications of common marketing analytics to problems such as: engagement, conversion, retention, churn, recommendations, forecasting, segmentation, experimentation, lifetime value, etc.What it does well is that it provides exposure to a wide range of applications of data science techniques to some pretty good datasets that somewhat replicate what you'd see out in the wild. The author also shows some data manipulation and aggregation techniques that I was not familiar with, like the use of "Groupers" combined with timestamp resampling inside of "group_by" operations. His notebooks are all neatly organized in the book's repo and the code is very readable.From here on out, the book is a bit of a letdown. The most salient of its shortcomings involves the depth (or lack thereof) of its exposition when it comes to the technique or algorithm being discussed. There are a couple of simple formulas here and there and very short explanations of what the algorithms are doing behind the scenes. I guess that's ok if it is meant to be an intro and not a full-fledged encyclopedia, however, I cannot overlook the sidestepping of common pitfalls and best-practices when it comes to leveraging these techniques in other places.For instance, in chapter 6 (collaborative filtering) the author walks the reader through the application of user-user and item-item collaborative filtering using unary matrices and cosine distances, without mention of assumptions made of the data (you need a lot), on the shortcoming of the technique (collaborative filtering doesn't work well for new users or new items), why you'd want one vs. the other (user-user CF is much more computationally expensive than item-item since one is more likely to have more users than items), or why you'd use cosine distance vs. Pearson (mean centering), etc. This is pretty much the case in each section.There are some editing quibbles as well. Chapters 5, 7, and 8 are pretty much rehashed from earlier chapters, so there's no reason why the material couldn't have been consolidated.Also, I wonder what the folks at Packt were thinking in packing essentially the same book twice into the same print - one for R and one for Python. If you are a Python user, then half of the book is dead-weight (the R chapters and vice-versa). Why not have "Hands-On Data Science for Marketing with Python" and "Hands-On Data Science for Marketing with R"?I have no issues recommending this book to students in an introductory class as long as there's some adult supervision. I'd be slightly concerned that the exposition nonchalance might lead some practitioners into thinking data science is just a couple of one-liners and boilerplate API calls...

If you are using this book for a class and need to use the book to do some python programming, get any other file format than kindle.Everytime you go to copy the python code into a notebook, you get 2 problems1) every single time you copy and past anything, it attaches this "Hwang, Yoon Hyup. Hands-On Data Science for Marketing: Improve your marketing strategies with machine learning using Python and R (p. 86). Packt Publishing. Kindle Edition."so you have to manually delete that EVERYTIME2) the code doesnt work!There are formatting errors that happen between copying and pasting, so even after paste it, you delete the additional sentence about your source, and try and run it, it wont work.I had to troubleshoot EVERY LINE OF CODE with chatgpt. Take my advice and do not even bother with the kindle version of this book.

It took me a while to find a textbook for a course I teach at Harvard on Data Science in Advertising Technology, so I'm glad I found this one. It's on Marketing, not Advertising, but it does a great job covering common business use cases for data science (e.g. customer acquisition; product recommendation; customer lifetime value) by introducing a new machine learning algorithm (e.g. logistic regression; decision trees; clustering) in each chapter. And the best part is, it includes all the code a student needs to explore the data, build the models, and visualize the results -- in both Python and R, the two most commonly used open-source data science languages.

Muy interesante todos los proyectos que se pueden hacer en el area de marketing, mos ejemlos muy claros para diferentes tipos de problemas reales.I refommend it.

The content is very well written and easy to read. The topics are covered with sufficient background to understand how to drive good causal inferences with hand-on training examples.

Perhaps I am a bit bias because I've read other Data Science books but it lacks color, page design, and overall the descriptions are not robust. It feels like a compilation of low quality posts from Towards Data Science. It seems Python and R portions should two different books, it's way too redundant. Not sure whether if I should return this or just power through the book.

This book comes paired with a GitHub repo with all the code. However, the code depends on the data which is nowhere to be found in the repo. I have no interest in going to 70 different links, downloading the data, and putting it in the right folder per chapter just to follow along. Huge waste of time, if I could give this a negative rating I would. Save yourself hours and go find another option. I genuinely cannot believe this was ever approved for publishing in this manner

The author does a comprehensive job of building knowledge frameworks to explain how analytics can be used for marketing datasets. Bonus stars on providing code for both R and Python. Covers a lot of ground that will be useful to brands operating in the retail, D2C space. Was personally looking to study the theory of attribution models and market mix models which was not sufficiently detailed in this edition. Hope the authors can address this in their subsequent revisions.

Perfect for someone who is looking for a data science starter kit in the field of marketing. Highly recommend.

Very practical book. Recommended marketers and data scientists working in marketing.

Es ist wieder so ein Buch das on-demand gedruckt wird. Schreckliche Qualität.