This could be an awesome command line SEO tool (feature requests)

[teomaragakis]

This project has a ton of potential as a simple, command line SEO tool. I think the following features would be really beneficial:

• Provide input files with keywords. Something like google -q file.csv (or .txt or whatever) and get results for a comma-separated list of keywords.
• Website rank mode. google -w www.somewebsite.com -q "query" (or -q input_file.txt). Outputs results for a specific website with a number ranking.
• Project mode. Use a local JSON file with a structure of your choice, combining the above (a website and a list of keywords) plus a list of TLDs and languages. Specified by input file: google -p mywebsite.json. Returns a nicely structured output for each language/TLD combination and for all the keywords. Could use the colors nom package (and/or others) for pretty output.

I'm not great with node, but if you like this idea and need help, I'll help in any way I can.

Also, I have been working on a frontend for a web-based open-source SEO tool, based on your package. If you are good with the backend stuff, we could work on it together. I'm thinking MVC with express, angularJS and mongoDB.

[jprichardson]

This project has a ton of potential as a simple, command line SEO tool. I think the following features would be really beneficial

I actually like a lot of these things, but I think they're best left to other packages that then would require('google'). However, I'm willing to modify this package to enable usages in your aforementioned suggestions. For example, there was a suggestion of having this module retrieve Google AdWords results. That may be within the scope of this package, but it could definitely at least return the contents of the body to enable the behavior.

Also, I have been working on a frontend for a web-based open-source SEO tool, based on your package. If you are good with the backend stuff, we could work on it together. I'm thinking MVC with express, angularJS and mongoDB.

I'm not sure what you mean by the backend here?

[teomaragakis]

I think that out of my 3 bullet points, #1 would be the only one which would perfectly fit the scope of your project. The others might be cool, but you are right, they are a level higher.

By the backend I mean the server code. I'm a views guy. I am a great with design, but my server-side stuff is more PHP and WordPress oriented, doubt I could pull off a great tool with node.

As far as implementation goes, I'm really liking the angular-fullstack generator for yeoman.

[jprichardson]

By the backend I mean the server code. I'm a views guy. I am a great with design, but my server-side stuff is more PHP and WordPress oriented, doubt I could pull off a great tool with node.

Sorry, what I meant was, what does the tool that you're working on do? What would the backend do?

[teomaragakis]

Sorry, what I meant was, what does the tool that you're working on do? What would the backend do?

Users would add websites and keywords they are interested in tracking (along with TDL and language).
Then they would see how their websites rank for these keywords and charts (to show how the ranking changes in time). Maybe get monthly searches for these keywords as well. That's roughly what I would imagine in a first version.

Then I suppose more stuff would be added, like page analysis, backlink tracking and more, but for the first version that would be more than enough.

[jprichardson]

Hmm, you may have something here. But it sounds to me that it may actually have a lot of viability as a commercial tool.

[teomaragakis]

It does, but prices of commercial apps are ridiculous and the UIs suck. I'd rather do a first version in open-source, then if it works well and you want to move it forward, we can definitely work on it. But a proof of concept is needed first.

I used to pay for an app like that, it was something like $10/month for 15 keywords.

In any case, add me on Skype (newholborn) and let's chat :)

[fbgranado]

HI,

First, i am sorry for my bad English.

Sometimes I try to make a seo tool with a friend. We used Sails(node) but we have a lot of problems. The major problem is the fact that Google block us when we make a lot of requests.

What we did:

• The user can set a main url and other urls ;
• The user set a keyword. We search this keyword and check if the results match any setted urls. (1 request)
• For any found url we try to get the inbound links. Sometimes google block us here.

I try to use the module async with the function eachSeries to make requests in an order and put setTimeout between the requests, but did't work. When google block, they block the IP.

att

[jprichardson]

@FeliPK you should have started a new issue.

I'm closing this issue as it's been tainted. @teomaragakis you can open another issue if you'd like.

@FeliPK I'm not interested in discussing how to get around Google blocking requests and don't want this project or Github repo to be associated as such. That is outside the scope of this module. Google has a right to block people from excessively using their resources and discussing methods or means to get around it won't be tolerated in this project.