Maximize your tweet's chance of discovery via hashtags | Hacker News Reader

Maximize your tweet's chance of discovery via hashtags

ritetag.com

36 pointsmaintopbiz42 comments

This is a cool idea, but can someone explain why this requires the following Twitter permissions?

- Update your profile.

- Post Tweets for you.

- Access your direct messages.

In general I'm at loss why so many apps require so many permissions. They must lose a ton of customers from this. Anyone care to explain? Is it just a spam app meant to create a botnet or what?

My thought (as a non techie) is that it's just easier to ask for everything.

Need to use an open source library in building your codebase? No need to analyse everything to see which permissions it does or doesn't impinge upon. Want to add a new feature down the track? No need to consider whether it changes the user permissions.

And I'd be surprised how many people are turned off by the number of permissions - I'll almost always back off if there are any, and if I really want to use the product I have to trust them to do no evil. My guess is an awful lot of people are always living in that trust space, especially if they don't understand the tech.

My thought (as a developer) is that it's just lazier (and worse for the product) to ask for everything. The more permissions you ask for, the less users you'll get.

For example, Facebook blatantly recommends that app developers ask for only the bare minimum of permissions on the initial app engagement. They cite numbers showing that click-through rate falls off precipitously once users see more than a few permissions required. They say that you should only ask for more permissions as the user explicitly tries to do something that requires them, for instance, ask for sharing rights only when they click 'Share', not before.

Lots of people are turned off by the number of permissions. Be surprised.

Yup. I think OAuth calls it scope, and it's a list of requested permissions.

Sure, it's a turn-off. However, people are learning that APIs and social network auth policies confine devs; it's a neccasary evil. We're blessed by a plethora of press that evidences people overlooking the Twitter auth thing and signing up nevertheless.

Hi, I am a co-founder of RiteTag. Don't worry we are not any spam bot. But unfurtunately, Twitter has only 3 types of permissions:

1) Read 2) Write 3) Direct messages

We cannot go only for Read because we need to allow users to sent and schedule tweets via RiteTag. But we do not send anything that user hasn't manually approved.

Write permission goes automatically with Update your profile even though we don't use it at all. As developers, we cannot select only Posting tweets. That's Twitter's policy. It doesn't make sense for us either.

Lastly, we were playing with direct messages a year ago. We don't need them now and we could turn them off. But in the meantime we got more than 7000 users and if we change the permissions RiteTag would stop working for them until they re-authorize it. This means all the tweets they have scheduled would not be sent.

Here is more info from Twitter, if you are interested: https://dev.twitter.com/docs/application-permission-model

Thank you for your detailed response. I understand that there are technical limitations when it comes to permission, but it still makes me uneasy when I have to give that many permissions to a new app. Maybe it's the granularity at Twitter that is wrong and should be changed, but if there's any way you can ease down on the permissions I think that would be appreciated by many users.

Regarding not wanting to turn off the DM permission in order to avoid a disruption to 7000 users, as the DM permission is a disquieting thing that will prompt complaints as this gains popularity and 7000 isn't that high a number, you might want to just rip the band aid off and get it over with now..

Founder of RiteTag here, and always available @osakasaul in Twitter (Saul Fleischman). FWIW, in regards to 7K users not being a high number, it took us 25 months to get to 7K users. Many signed up well over a year ago. Asking them all to remove permission in Twitter settings and then auth in again, we'd lose so many people who are not used to going into Twitter settings, and would simply go away. But thank you very much for the feedback - and also, a worthwhile opinion on the problem with Twitter permissions.

After I saw "Access your direct messages." I cancelled my signup & uninstalled the browser extension. That permission is a bit too steep for me.

Sorry to hear that. We don't really use this permission at all. See my explanation above.

FWIW, founder here, and let me offer you a personal tour. I'd be happy to screen-share, show you my own RiteTag account, my stats, what I do with RiteTag and hashtags in general. I'm sure our team would benefit from your thoughts. Thanks, in advance.

I played with twitter bots before, and I tried different ways to maximize the exposure of my "automated" tweets, sometimes by parsing dictionary words, other times by targeting trending hashtags, but that didn't seem to generate much traffic.

I would be really interested in any statistics regarding the effect of hashtags, or statistics about users who actually browse/search through hashtags.

I submitted a post yesterday I wrote about using statistical analysis on Instagram hashtags: https://news.ycombinator.com/item?id=7458540

The key takeaway is that there's a lot of variance in the amount of engagement even after optimizing your hashtags. Twitter would be worse since there's an implicit hashtag cap.

Can I give you some unsolicited feedback?

- On Firefox 27x the "Most Popular" note is overlapped on top of the pricing, making both hard to read

- the jump from free to lite seems a little large... is there no room for, say, a $5/mo plan?

- after authorizing Twitter in order to sign up, the site blanked out and I had to go back to ritetag.com manually, after which I was logged in

- I can't tell a difference between the colors for "tweets with low chance" and "tweets with high chance". I'm assuming one is orange and one is green, but like a sizable minority of your users, I'm essentially unable to distinguish between the two effectively given the shades you've chosen.

- Related, "tweets with a high chance of being discovered" wraps to the far left, instead of just below (so it looks like on the other side we have "tweets with no chance of being discovered discovered"

Otherwise, I'm looking forward to exploring more -- interesting concept, and I personally haven't seen this done before.

Thanks a lot! We will check the Firefox issues and learn more about color-blindness to make RiteTag more accessible.

Regarding the cheaper plan, we'll need to think it through as small amounts are not profitable due fixed costs per transaction.

Obligatory Bitcoin reference... Check out integrating with Coinbase?

Founder, here. Few ask for bitcoin for payment, but the bigger truth is that when we need to show investors that people see value, we actually can't risk being told "right. people paid you with something that cost them nada to acquire." This is not my stance on bitcoin, but we do need to respect the big wigs' decision-leading standpoints.

Be sure you save that great comment, Mr. CTO. That's why we're in Hacker News; the people who help us see where we're in need of improvement. (Thanks, @binxbolling!)

In regards to plan cheaper than our "Lite" plan, kindly keep in mind that we have hard costs. Scaling up to meet the needs of API customers, and also, recaculations on the nearly 2 million Hashtag Scans that our users reply on... recalculations on millions of hashtags as for their likelihood of leading to discovery, this take a lot of access. Further, we provide actual analytics that help RiteTaggers learn from their efforts. I don't believe you'll find anything nearly as cheap as our Pro plan - let alone the Lite plan - that does anything close to this. When you're in there and do some smart-tagged tweets from RiteTag, you'll see this in Stats and also My Top-performing tags (Tweet Composer). But do let me know @osakasaul if you still believe we're costly.

Do people actually watch hashtags that aren't event specific?

yes, social media mavens and SEO gurus ;)

My thoughts exactly. They always look a bit #desperate #tryingtoohard #inauthentic PLEASE RT!!!

That would be a funny contest: most desperate tweet.

Okay, when people want to be #ironic or don't care about reaching beyond their followers, we probably don't have much for them. Still, if they want to track hashtags associated with a topic, product, event, brand or hashtag they know, our Alerts are a must.

But for those actually making money or waves in Twitter, people actually want to stop guessing, and go with our grading.

I scroll through #d3js every few days to see what's being made/talked about.

Yes, and many people set up a column in Hotsuite/Tweetdeck/etc. to track a few hashtags passively. When someone's using #ritetag or #startuplife, #hashtagmarketing, I see it this way.

I've recently joined RiteTag on their PR/Marketing team, and I understand where you're coming from. Many users on Twitter use it primarily to connect with friends, however, there are also countless users that work in PR/Marketing and other sectors that would have a need to connect with their customers as well as reaching to a larger audience. Since you can also search hashtags now on Google, keeping track of hashtags related to your topic of interest can be helpful!

I thought this was a joke then I came to the comments and everyone is being serious. I guess they've changed what "it" is.

You can also look up the hashtag and check traffic there. If there is a tweet every 2s => too much traffic to get discovered. If there is a tweet only every 30min => chance high. Don't get why I would pay $15+ for such a service.

One of our goals is to make it easy for our users. It will reduce the amount of time spent on evaluating hashtags and guesswork. We are also integrated in places other than Twitter, such as Hootsuite, SocialOomph and Buffer. We have functions other than checking hashtag traffic where we allow you to schedule tweets, check for associated hashtags, and many more.

You can try out our service for free first to further evaluate the value of RiteTag. I hope you will reconsider!

I've tried to download the browser extension and got 404 error, tried to find the app in google store and no luck. Is it geographically limited? What can be the reason? Thanks.

Hi. Here is the link:https://chrome.google.com/webstore/detail/ritetag-find-the-b...

We'll check the links.

This sounded really helpful, then I tried to use it and ended up with this: http://i.imgur.com/LsPHVd6.png

Derp.

RiteTag shows associated hashtags that are often used with your query. We didn't make that up. That's what people really use.

Does this account how often people search for each hashtag? Do people really discover tweets with generic hashtags like #Guide?

There is personal experience - from our people and the many that talk about hashtag successes. As examples, "operation wallstreet" rallied tremendous community support with #ows and then related hashtags like #owsboston etc. When there was media blackout about a rigged election in Iran and phonelines were even blocked, a hashtag spread the truth. What's more, it was my personal experience, during the big earthquake and tsunami (tidal wave) in Japan, and seeing how I could find very relevant tweets from hashtag streams rather than the cover-up on Fukushima on TV that led me to understand that hashtags have power - but that no one knew which was any better than another.

I clicked around a few pages and couldn't see anywhere which explained what the different colours mean.

I apologize for the inconvenience, and thank you for the feedback. I'll link you to two of our blogposts that explain a lot of the RiteTag functions.

Pink is "overused", Blue is "good" and Green is "Great". The blogposts will explain more of our functionality!

http://ritetag.com/hashtag-academy/workshop-2 http://ritetag.com/hashtag-academy/hashtag-virality-measurem...

Wow cool!

Would really like to see related tags, built in the usual big data algorithms / search UX way.

We'll soon release an API if someone wants to play with hashtags ;-) http://ritetag.com/rest-api

What? I thought this was Hacker News, not Marketing News.

Actually, hashtags are not just for marketing, nor is RiteTag. People get plenty done, such as bringing people to #hackforgood or hackathon events, search for lost children, and more. But without a hashtag grading system, they really have no idea which hashtags are best for getting discovered.

Reply on news.ycombinator.com