Tell HN: I released my open source iPhone AppStore Sales Graphing Tool | Hacker News Reader

Tell HN: I released my open source iPhone AppStore Sales Graphing Tool

maxklein.github.com

38 pointsmaxklein14 comments

I released a GAE app which does something similar. However it differs in that it is automated and it emails you a daily digest of your sales, upgrades and current rankings. In the email are also graphs of your sale and upgrade history. I've been using it successfully for several months now. It's also written in Python and on GitHub. :) http://github.com/baz/app-sales-machine

That's pretty cool. I'll see if there are parts from your project that I can steal. You know what I'd love your project to do? To post the sales report to google wave. I wonder how hard that would be. Can you estimate?

I think you would have to build a Wave Gadget. Would be pretty cool because then you would have the functionality of the playback scrubber to see what your sales graphs looked like at any point in time.

It would probably take a couple of days to open up a service on the server-side, and to write the client-side graph generation. Not sure if a Wave Gadget can pre-emptively fetch data without user interaction though. That might be a killer.

Can wave gadgets not render standard html? I never thought of the scrubber thing, but you're right, it would be pretty nifty.

Is it not possible to simply push the current html you have into a wave?

Wave Gadgets are HTML/JS. I was concerned that an external source (i.e. App Sales Machine) would not be able to affect the contents of a Wave Gadget unless there was some direct user-interaction with the Gadget inside the Wave.

Any experience Wave Gadget builders want to chime in? Is this possible?

I just looked it up: http://code.google.com/p/imdbotty/source/browse/trunk/gadget...

It seems that the html is server-side preprocessed before being displayed in the gadget, if I am correct in assuming those are template tags. So that would not be a limitation.

I wanted to install your tool, but it is an inconvenient tool to use if you a lot of apps. Why don't you just pull the app ids out of the sales file and generate the dictionary that way? Tnen it would work well with multiple apps.

App Sales Machine also scrapes your app's rankings, so that config file is used to generate the URLs to scrape. From there, I relied on it to help organise the way the data is persisted in the data store. I suppose it could be modified to parse everything it needed from the sales reports, but you would still need to configure every app's category info for the rankings scraper. I'm sure there are other ways of doing it, but this is the method I settled on.

It is trivial to have it send to gTalk too. I have a BI app on AppEngine for a couple of retail stores and get daily sales figures delivered via gTalk nightly.

Thank you for sharing this. Both projects look really cool, but I do a lot of AppEngine development already and need to share the reports with a business partner.

Kudos to both projects though. Definitely fills a need.

Also, have you guys seen the sample reports from ZohoReports for AppStore sales? The missing link is updating the data, which it seems both of these projects do.

http://blogs.zoho.com/general/analyzing-iphone-app-sales-wit...

Cool. I'm not that familiar with Zoho's products, but if there was an API available to allow pushing new CSV reports, you'd be set. Just set up a GAE cron job to run every morning.

Actually, Zoho will pull from a web accessible CSV. I don't know if you can use their reports directly, but if so the GAE parser could add another model to capture all fields from the report and make the daily CSV available. Then the Zoho Reports would update automatically.

Add in some new reports for rankings and you have quite an App Store dashboard.

Nice. In fact App Sales Machine can already generate a CSV report. The problem then may be circumventing Google Account authorisation without making your data public.

At the bottom of the site I've listed a bunch of things I'm planning for the future. If anyone can come up with ideas on what would make for a really good sales grapher, please do comment.

Looks nice, max. I thought I'd finally be able to see what some of your apps were but alas, no!

Reply on news.ycombinator.com