Hacker News new | comments | show | ask | jobs | submit login
The Horizontal Distribution Hack (growhack.com)
25 points by conradwa 1368 days ago | hide | past | web | 17 comments | favorite



I think this is a unique observation that doesn't really scale to other platforms: Facebook is a giant sea of rubbish apps, and the goal is to snatch attention from someone scrolling through a whole pile of their friends' updates. In this case it makes sense to post something as appealing and specific as possible to the timeline, to better appeal to a random passerby. App discovery on mobile is not drive-by: none of these apps will ever hit a leaderboard (splitting them up fragments the userbase), and they'll look bad in search results because of few downloads, few reviews, and a seemingly over-specific purpose.

That's not to mention the fact that Apple won't let you publish 200,000 apps from one account. This is like an article recommending that everyone become a bounty hunter because it was a good way to make money in the Old West: it takes advantage of a single market at a specific point in time. I wouldn't be surprised if Facebook starts cracking down on this sort of spam as well soon.


This guy's Facebook tactics haven't been valid for over 4 years and they certainly won't work to the same degree on the App Store.

This particular tactic worked on Facebook because ratings of spaminness and blocking happened on a per-app basis. You could take a spammy app, reskin it 1,000 ways, and release them in to the wild seperately. If a user blocked "Diwali gifts" app they still get messages from "Christmas gifts" app. It's essentially a hack to workaround the way Facebook's app ecosystem was designed at the time.

He has two really important points, whether he knows it or not -

* There is a powerful long-tail search of apps in the App Store

* Users will be more likely to click and download an app that is more targeted at their search intent

Both of these are really hard to leverage with a single app right now in the App Store. Apple WILL shut you down if you create 10,000 shallow copies of your app, so that isn't an option.


I have not developed for the Facebook platform since the latter half of 2011 so there might be some truth in it - but this was certainly effective up until then.

The spam-rating was only one aspect of it - horizontal distribution attracted users that otherwise simply would not have been interested in the application otherwise.

Regarding iOS shutting you down, I think two things are important. The first is making a good case for the application not being shallow - looking at the way Udemy has full course (some of which have 10+ hours of material) on drastically different topics as separate applications might be a good template. They have over 100 iOS Apps and have not faced any action.

The other interesting strategy to consider (one that I haven't tested) is talking power users (such as meetup group organizers) through the process of generating the correct signing keys and then letting them download a complete application binary that they independently submit to the App Store under their own account. This casts a wider net and potentially might alleviate the getting shut down issue.

With that said, I haven't tested it on iOS myself so cannot speak authoritatively.

I have succeeded with horizontal distribution on Google Play by using screen automation technology to submit applications - but finally, had to take it down since the applications were based on quotes from TV shows which Google claimed did not come under the Fair Use Act of the Copyright law (whole other issue)


Nice. I call this technique the "attack of the clones". I was slammed with this method by competitors in FB apps in 2007.

Cleverly done my nemesis. :) As you note, this is effective in many other places too.


The premise here doesn't support the hypothesis. We start out talking about Facebook apps, and it makes sense. The name of a game, when it appears on a timeline, is far more compelling than the brand names. If you're trying to sell a product, you need to get the product in front.

Here's where things start to break down. People don't look for meetups in their apps - they look for meetups online. No one looks for a "SF singles app." They don't even go to the appstore for that - they go to Google. And despite the fact that I pulled this term out of a hat, guess who's first in Google for "SF singles meetup"? Meetup.com.

The second reason this doesn't apply - meetup isn't selling a SF singles app like a company sells a game. They're selling a platform to help people organize. By putting the app name in front, they put the product in back.

Honestly I think the fragmentation of apps would be a mess, and that it would ultimately result in lower visibility, awareness, and traffic figures for meetup.com as the meetups themselves become the brand.

You cannoy simply assume that a branding strategy/architecture that worked in one case will work as well for a different product or industry. Sometimes umbrella/family branding is the way to go.


You're right, and the reason why you're right is pretty interesting. Meetup.com's business model is built around charging meetup organizers and building the relationships between organizers and participants. So Meetup's interest is to form a lasting relationship with it's users with the aim of converting them to organizers; and single purpose distribution of apps for individual meetups doesn't help that.

I would say that the OP is looking at the wrong metric for meetup's business model; it's not mobile traffic that counts, it's contacts per user per month[1] that tells meetup whether they're succeeding at building lasting relationships.

1. This would actually be a weighted moving average of the last three months and another of the past year. Obviously, not the only metric in play.


I don't think the argument is focused on what Meetup should do, but what they could do if new user acquisition were a priority. It's a growth tactic that can be balanced with retention and engagement approaches, that Meetup right now understands and is heavily investing in.

Taking this approach also doesn't mean you'd have to go full throttle. You could test and decrease the pace by only allowing a segment of quality organizers the ability to create their own app.


Nothing wrong with an umbrella approach - I'd argue it's likely that Meetup cares a lot more about retention and engagement, and cross-selling other high-quality events, than any type of acquisition at this point. Users bring in other users through word of mouth is a pretty compelling approach.

That being said, if they wanted to open up their new user acquisition efforts on mobile they could let users brand and launch their own applications in addition to their main app.

They could do it in the same way you're becoming more able to customize a meetup landing page now, and tuck promotion of other similar events, or even the main meetup app in the background.

This type of approach could appeal to organizer who really cares about their own brand. Think large conferences and related topics, who'd love to use a Meetup backend, but need to differentiate their brand.

On the search point would you agree that someone searching for "soccer" in the iOS store would be interested in an app dedicated to doing soccer in person?


(Disclaimer: I am the author of the article)

I disagree - I believe people have an intimate enough relationship with certain meetup groups that they work as a micro-social network. To that end, it would make logical sense to have an independent application for those groups.

With that said, this strategy would not be an alternative to having a central Meetup.com application - this strategy would work in conjunction primarily to leverage the additional app store SEO boost from having thousands of unique titles. Each individual application would link right back to Meetup.com's primary application and people would have to sign up for a Meetup.com account in order to use it.

With that said, there are shortcomings to this approach. This empowers group organizers a little more than Meetup.com would ideally like and might work better for companies that white-label their services.

I still don't doubt that it will greatly increase raw traffic numbers - but maybe thats not what their priority is at the moment.


>"I disagree - I believe people have an intimate enough relationship with certain meetup groups that they work as a micro-social network. To that end, it would make logical sense to have an independent application for those groups."

But what does that do for meetup.com, except cut themselves off completely from the very communities they're trying to foster?

>"to leverage the additional app store SEO boost from having thousands of unique titles."

You only get the "SEO boost" if people are actually looking for terms on a search engine. You're still assuming that people are actually looking for local interest group on the app store. I really don't think so.

If anyone were serious about it, one could run a quick survey and ask where users would search if they wanted to find a local group to talk about shared interests. I'd be surprised if more than 1 in 500 said anything like "app store".

>" still don't doubt that it will greatly increase raw traffic numbers"

I feel like a jerk, now, and I don't mean to be, but I also have to disagree here. I don't see how cutting Meetup out of the equation amounts to more traffic for Meetup.com.


I think the fundamental difference between our points of view is that I am attributing traffic on each individual application to be contributing to Meetup.com's traffic.

Each individual application is still owned by Meetup, it runs their application binary executing their advertisements, running on their payment processing, cross-promoting whatever content it is that they want to promote - what difference does it make if the application is called New York Tech instead of Meetup.com?


I think one difference is that they will recommend New York Tech to their tech friends, rather than recommend a general system for meeting people with shared interests to all of their friends.


As some have mentioned, this is essentially spamming the markets with shallow copies of your apps, which is typically verboten by major app stores. What's really necessary to truly take advantage of this concept without being spammy is a better form of keyword association for apps, allowing you to associate a gift app with "birthday", "holiday", "diwali", "christmas", "hannukah", and all the other holidays and occasions I can't think of at the moment. Of course this would need some sort of check to prevent people from spamming keywords you're not really relevant to (maybe a means of downvoting one of the keywords that brought you to the app on the app's page?), but assuming you can enforce a degree of responsible usage, this would allow app makers to build out the semantic associations of what their app can provide to people. Google in particular ought to be interested in something like this, as it would allow them to further build out their sematic trees, and associate holiday gift apps with specific holiday keywords more easily, enhancing their search results and ensuring they stay ahead in the app-market game.


Let the app creator create initial tags. Have system tags like spam or malware, that cause the app to be reviewed. Let users add and vote up or down tags, perhaps prompting them when they update (so they will have used the app. Downvoted tags are removed. Upvoted tags are placed higher in a search for that. let users searching specify tag only, or app name and description, or both.


So, in short, spam the app stores? ;)


This makes a lot of sense. Someone should be an easy GUI to create lightweight Facebook apps like the Meetup apps the author talks about. (Maybe this exists. I've never developed for Facebook.)


Meetup already integrates with Facebook, and it shows the name of the Meetup rather than the product. Meetup wouldn't gain any additional advantage from breaking their app down into a million pieces.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: