Andrea Volpini Profile picture
Jul 5, 2022 โ€ข 5 tweets โ€ข 3 min read โ€ข Read on X
This is great, and here is what you can do with it ๐Ÿงต:

1. Blend structured data with metrics from Google Search Console ๐Ÿ”€(or any other source).

2. Evaluate clicks performance for each schema type
(including types like schema:service that we currently don't have in GSC).

1/5 Image
3. Filter data by entity (or topic). Make sure to activate the Chart cross-filtering option to turn any chart into an interactive filter.

2/5 Image
4. Find query opportunities using Google's Search Console bubble chart (ref: developers.google.com/search/blog/20โ€ฆ) and entities.

For example, I see everything in the context of "Smart City" ๐Ÿ‘‡.

3/5 Image
5. If you run an e-commerce ๐Ÿ›’ website, you can analyze sessions by price and product category.

6. You can see what brand bring you the most visits.

(in both of these charts, the data from the KG is blended with data from #GoogleAnalytics)

4/5 Image
Here is how it works:

1โƒฃ Add your @wordliftit key and write a query in GraphQL with the data points you need. The connector will access the KG and flatten the data into dimensions.

2โƒฃ Create a blend by matching the URL (products.urls ๐Ÿ”— Landing Page)

3โƒฃ Have fun!

5/5 ImageImageImage

โ€ข โ€ข โ€ข

Missing some Tweet in this thread? You can try to force a refresh
ใ€€

Keep Current with Andrea Volpini

Andrea Volpini Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @cyberandy

Jul 24, 2023
๐Ÿงต ๐—˜๐—บ๐—ฏ๐—ฒ๐—ฑ๐—ฑ๐—ถ๐—ป๐—ด๐˜€ & ๐—ฆ๐—˜๐—ข: A Powerful Multitool ๐Ÿ› ๏ธ

In the world of #SEO, understanding and leveraging the power of #embeddings is a game-changer. But what are embeddings, and how do they work? Can we build our embeddings?

Let's dive in! #AI #Semantics

1/9 Image
๐—ช๐—ต๐—ฎ๐˜ ๐—ฎ๐—ฟ๐—ฒ ๐—ช๐—ผ๐—ฟ๐—ฑ ๐—˜๐—บ๐—ฏ๐—ฒ๐—ฑ๐—ฑ๐—ถ๐—ป๐—ด๐˜€? ๐Ÿ“š
Word embeddings are a type of word representation that allows words with similar meanings to have a similar representation in a multi-dimensional space. Using embeddings, we convert words into math (numerical vectors).

2/9 Image
๐—›๐—ผ๐˜„ ๐—ฑ๐—ผ ๐—˜๐—บ๐—ฏ๐—ฒ๐—ฑ๐—ฑ๐—ถ๐—ป๐—ด๐˜€ ๐—ช๐—ผ๐—ฟ๐—ธ? ๐Ÿง 
Embeddings map words into a high-dimensional space where the distance and direction between words indicate their semantic relationship. Words with similar meanings are closer together, while different words are further apart.

3/9
Read 9 tweets
Jun 12, 2023
๐ŸŽบNew Article on Neuro-Symbolic AI and #SEO๐ŸŽบ

๐Ÿ’ก Neuro-Symbolic AI is a way to combine the power of knowledge graphs with deep learning techniques using *Retrieval Augmented Generation*.

๐Ÿ”— Article: wordlift.io/blog/en/neuro-โ€ฆ
๐Ÿ’ฌ AskMe: a widget inside the article

1/4 ๐Ÿงต
๐Ÿ’ฌ AskMe is the first experiment to let readers "talk" with the article's author and ask questions.

โš ๏ธ It is a WIP and requires your patience ๐ŸŒ

It has been developed by the wonderful @wordliftit team using the KG behind the blog, @llama_index, @LangChainAI and @JinaAI_

2/4
๐Ÿ’ฌ AskMe is using in this implementation:

1โƒฃ The latest 10 articles I wrote on the blog.
2โƒฃ My entity page (as *system message*).
3โƒฃThe article the user is reading.
4โƒฃA combination of the two indices above (a graph index).

3/4
Read 4 tweets
Mar 27, 2023
I worked on my first experiment with ChatGPT Plugins applied to SEO! I created a search engine for an eCommerce site using WordLift's Product Knowledge Graph (PKG) as a starting point. Here is what I learned:

#SEO #ChatGPT #WordLift #eCommerce

๐Ÿงต 1/6
1. You can use @LangChainAI to experiment with the idea (here is the documentation ๐Ÿ“–: python.langchain.com/en/latest/moduโ€ฆ) and choose any of the supported LLM. #Opensource is amazing ๐Ÿคฉ.

2. You only need two files: a JSON manifest and the OpenAPI specs.

2/6
I built the search engine using @JinaAI_ by indexing schema:Product in the PKG WordLift created by accessing the Google Merchant APIs.

๐Ÿ”Ž nowrun.jina.ai/?host=grpcs://โ€ฆ (you can play with the search engine here)

3. You need all your prompting skill to create the JSON manifest.

3/6
Read 6 tweets
Mar 2, 2023
๐Ÿ“ข AI-assisted Alt Text for Images ๐Ÿช„โœจ

๐ŸงตImage SEO has tremendous potentials as AI-powered technologies make search more visual and intuitive.

โœ… wordlift.io/blog/en/image-โ€ฆ [blog post]
๐Ÿค– wor.ai/image-captioniโ€ฆ [Colab]

Interested in automatic captioning? This is for you

1/
Google has recently updated the Image SEO best practices, and if you missed it, I suggest you start from @RoxanaStingu's update:



I have worked on automatic image captioning since 2019; recently, I have done a new analysis of the SOTA.

2/ Image
I am generally less interested in models and more interested in finding the framework to fine-tune and evaluate models, so I found LAVIS by @salesforce, which provides access to 30 pre-trained and task-specific fine-tuned models checkpoints (ALBEF, BLIP, CLIP, and ALPRO).

3/
Read 8 tweets
Jan 24, 2023
A small thread about Semantic SEO and #ChatGPT ๐Ÿชก

As @SilviaKaya22 was testing @wordliftit new version of the SEO Add-on for Google Sheetsโ„ข, she quickly compared @aleyda and me ๐Ÿ˜ณ.

1/7
I feel like a newbie and an apprentice compared to Aleyda; nevertheless, the results grabbed my attention.

Using the new Add-on, we can compare search queries (and URLs) by looking at the entities behind top search results (or the entities behind single or multiple URLs).

2/7
We begin with two queries (Aleyda Solis and Andrea Volpini) and let the Add-on analyze the respective SERPs to highlight similarities and differences.

We both share, for example the entities:

- SEO (duh)
- entrepreneur (this also makes sense)

3/7
Read 7 tweets
Jun 9, 2022
A new SEO workflow for e-commerce sites. In this tutorial, we will create internal links on PLP

1โƒฃ Extract text using #advertools ๐Ÿ•ท๏ธ
2โƒฃ Get queries from GSC
3โƒฃ Compute Similarity
4โƒฃ Add semantic markup

- wordlift.io/blog/en/internโ€ฆ [blog postโœ๏ธ]
- bit.ly/3aRvN4o [code๐Ÿ]
I worked on the @FILAUSA website (not a client of ours) that currently doesn't have a related search widget.

To build one, we will index 'title', 'meta description', 'intro text' and 'breadcrumbs' and use the title of any other PLP as 'query'. ImageImage
We will generate a JSON that recommends 3 or 4 relevant links for every page.
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(