Market Search Tool

Getting ready to launch Market Search Tool app and would really really appreciate if anyone could help try to break it.

Market Search Tool is the first and only app to show what people are searching for on Google Play. Like wordtracker but for android. A must have for developers.

First phase will be primarily for operation, suggestions, bugs and pricing.

Second phase will be testing buying fake subscriptions and expiring thereof. This creates a significant server load so wiil be monitoring for how many more servers we will need to handle the load.

Third will be the final product with subscriptions enabled to buy with real money.

Just load url from your device to download and install. This version is the free version which shows only 5 results per query. The subscription (phase 2) will show 100 per query.

http://www.marketsearchtool.com/

Thank you so much.

It would be a great tool ,
However I have tried it and it giving me an error “Server is not responding , try again later”

Problem fixed, was returning that message when results = 0. Please download again.

Thank you Pharaoh

I get a fatal error. JNI not available

Ok, thank you cLin.

I have uploaded a new version, test subscriptions enabled now.

http://www.marketsearchtool.com/

Hi,

This app is now on the market. Thank you to those who helped out.

Hi Market Search Tool,
I’ve been meaning to tell you - when I use the app, it sometimes struggles to display the bottom advertisement. Hope this is helpful.

Ok, thanks Rick. Mucho appreciado.

Hi,
I have a question re your methodology - how does it work? On one hand you are not able to actually track real searches, on other hand you claim to track “actual searches performed on Google Play” (on your site). The idea is really good, don’t get me wrong, but it would be nice to understand how do you guys actually come up with these stats.
Cheers.

Hi Tom,

Every search term you see on the app was logged when a user searched for that term on their device on GP. We have software on millions of phones and tablets worldwide that record the searches and send it to our servers. (yes, it’s possible) We log about 100,000 real searches per day, give or take.

“On one hand you are not able to actually track real searches” Yes we do, every term in our app is a search made by a real human that we captured.

What we cannot do is tell how many time a specific search was made worldwide since we don’t have our software on every single phone in the world. Only Google can do that, so we have to estimate that total number. After a just few thousand searches this becomes very accurate.

For example, if we captured ‘fruit ninja’ being searched for 200k times last month through our software, we know for a fact there were many more of those searches on the devices that don’t have our software on it. So then we need to estimate how many times ‘fruit ninja’ was searched for worldwide by comparing how many times we got ‘facebook’ searched for as compared to what google adwords tool says is the total worldwide number is for ‘facebook’ and multiply that percentage diff to arrive at 65 million.

Actually, the search numbers don’t tell us much anyway. What’s important is what searches are being made and the popularity of one search over another. This is, in my opinion, crucial to app developers looking to find out which search terms to include in their titles and descriptions. Developers are restricted to typing in their term in GP and looking at the drop down with the search suggestions to see what’s popular, all 5 of them. And by the way, Google does not always show the top terms in the drop down, nor do they show misspellings.

:cool:

Thanks for the explanation. Calculating estimates is the way to go here and it would be difficult, if not impossible, to get better quality data. Let me dig a bit deeper :slight_smile: What is the software you have installed on users devices? I’m not asking for any kind of trade secrets but it would be good to understand the data flow.
Don’t get me wrong, I’m also in the data collection / crunching business and it is crucial to be as transparent as possible when it comes to how data is being collected.
Your concept is really clever and I’d be willing to pay a subscription fee.

The software is bundled with popular apps. The apps then communicate with the software in the backround and send the data to our servers periodically whenever the apps are used.

Hi Market Search Tool,

Another thing I noticed about the app is that when I do a search on a phrase, say “number converter” I get a result of a certain amount of about 5,000. But when I accidentally put a space after that phrase, "number converter ", it changes the results significantly, usually with no results. You might want to think about nixing any spaces after or before a word or phrase.

Rick

Hi Rick,

This is on purpose. This is because in the ‘starts with’ mode it is not searching for whole words but for exactly what you entered plus anything after the last letter. So if you search for ‘bird’, it will pull anything that starts with ‘bird’ including ‘birds’ and ‘birdbrain’. If you put a space at the end then it’s searching for 'bird ’ + anything after the space.

The ‘contains’ mode is the opposite, it searches for whole words only and will ignore the ending/starting spaces.

I think its a great idea and your methodology definitely has validity. Hell even using just the Google Keyword Tool itself works.

However, there is one caveat I have- the “Google Play to Google” ratio isn’t going to be perfect by any means. If there are 10x Google searches for FB it doesn’t mean there will be 10x searches for other keywords. It might be close some or even most of the time. But I’m sure there’s a lot of obscure popular keywords on Play that don’t hit or surpass that ratio.

Having said that, it’s still a great idea, but one has to keep that in mind.

Like you said, the numbers are nice but the relativity of keywords is much more important. And you’ve collected enough data to show that so that’s awesome.

Which brings me to my next point.

You should totally have another column that shows the top search term always be 1.000.

If the second search term has half the traffic, let that be 0.500, and so on. Or you can do percents, use 100 as the base.

Your app already does tell you this already indirectly through the downloads, but it’d nice to have a nice quick reference of relativity to look at.

You could call the two columns “True Relative Ratio” and “Estimated Total Downloads”

Just a suggestion for a feature I’d love.

Great app!

Hi,

Thnks for the kind words.

Yes, that is true, for example if we got a garbage search like “fdgh dfgh” only once(get that all the time). Then multiplying that would give a false result since it’s fair to assume that search is unique. However, this becomes much more accurate with highly searched terms. The best way would be to drop the multiplier altogether which is what should have been done.

As to your suggestion I have to laugh because that’s how it was originally designed. It was dropped because total searches would give more info and was more intuitive, and so the percentages became redundant, and not enough space for 2 columns anyway. I could easily include that feature as an option but unfortunately nobody is using this app, maybe 4 total. So at the end of this month the server will be cancelled and all subscriptions refunded and the app unpublished.

Made the mistake of making an app in a category that nobody searches for for the last time.

:frowning:

Hey Market Search Tool,

I was going to offer some other suggestions and also concur with Dan, Great App!

But I’m really disappointed that you’re going to remove it from the market. :frowning:

Is there any other title that you can release it under?

That’s a shame, because this app really has some great info it! :frowning: