Git Product home page Git Product logo

Comments (6)

pabloduque0 avatar pabloduque0 commented on May 2, 2024

Hey Mario!

Your article looks amazing 🎉! Thanks for sharing it with us.

Let us know if something seem unintuitive or had trouble finding documentation/example about something specific (we are working on adding more across the board but in case there was any thing specific).

Cheers!

from lightweight_mmm.

ledmaster avatar ledmaster commented on May 2, 2024

Thanks, Pablo!

I hope this encourages more people to try the tool.

The only thing that was not very clear was what to do with the organic data.

Is my thought correct that we don't need to add it as a media channel?

My models were slightly more accurate out-of-sample when I added it with cost=1 for everything, but it was not significant enough for me to trust it.

from lightweight_mmm.

pabloduque0 avatar pabloduque0 commented on May 2, 2024

Hello Mario!

Apologies for the delay in the response.

Good question! Could you give us a few more pointers on the nature of this organic data? There might be nuances depending on how it is generated or where it comes from.

That said, let me give a couple pointers on how that can be added (or not), here are some options:

  • Do not use it. If it is not very relevant or traffic is very minor.
  • Use it as a paid media channel (similar to what you did in your attempt to add it). You can introduce a fake cost for it (eg. average of costs of other channels, or average cost of a channel that is similar).
  • Use it as a non media variable. In LMMM you can add extra_features for the model to incorporate, this could be promotions that you run, competitors data, inflation data, ... but one could use organic data in certain scenarios.

I might be able to correct the given answer (or add on it) based on your response.

Let me know if something is unclear.

from lightweight_mmm.

pabloduque0 avatar pabloduque0 commented on May 2, 2024

I have also added a Community Spotlight section in the README and linked your article.

from lightweight_mmm.

ledmaster avatar ledmaster commented on May 2, 2024

Thanks for the highlight!

My organic traffic data (impressions) comes from Youtube and Google Search Console. Youtube is the largest by far.

I will try the extra_features approach (I completely forgot about it).

Everything is clear for now.

I am currently testing a few changes to my campaigns based on running the optimizer over the model with Audiences, Devices, and Demographics as channels.

The results make intuitive sense, so let's wait for the data :)

from lightweight_mmm.

pabloduque0 avatar pabloduque0 commented on May 2, 2024

Sounds great Mario! I would love to hear about the results so please feel free to reach out to me directly if you want.

Let us know if you run into any other problems with the library.

Closing this one but feel free to re-open.

from lightweight_mmm.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.