How to get actionable leads from Twitter in real time
blog.taskulu.com
blog.taskulu.com
I run a bot that posts Street Art pictures from Instagram [1]. The bot dominates the hashtag search term results [2]. People who find the bot, click the profile and click the URL associated with an app I made for discovering street art [3].
The bot dominates search results because it finds hashtag heavy images via Instagram. The hashtags on the Instagram image carry over to the Twitter post, resulting in an active searchable Twitter feed. The bot posts a over a thousand times a day.
Since releasing the app, I've averaged about 500 downloads per month, without any promotion. It works very well for consistent passive traffic.
[1]: Bot http://twitter.com/publicartfound
[2]: Search term https://twitter.com/search?q=streetart
[3]: Landing page http://publicart.io
If you watch some technology related keyword with search, anytime some stories from that keyword hit the tech blogs, you get a spam storm of hundreds or thousands of identical, useless tweets linking the post. I guess I'm being presumptuous, but that looks like they aren't doing anything at all to limit activity like that.
That said, I wouldn't say that it's in their BEST interest– too much spam and real users start getting turned off. So spam up-to-a-point.
What's the meaning of src=typd ?
I ran a series of experiments via http://newpublicartfoundation.com
This is your problem. Right there.
When I do ask, it's because I want responses from the people I follow (and that follow me) - trusted contacts whose opinions I am interested in. I don't want to opt in to a brand new spam feed every time I tweet.
"Trying to have a private conversation with friends? Have some advertising shoved in your face! Don't like it? GTFO!"
No wonder Twitter is dying.
Anyway, I don't use Twitter much anymore, but have to agree with the OP. If I wanted the opinion of random strangers, I wouldn't go to Twitter, I'd just Google it.
When it comes to communications technology, what is possible is not the same thing as what is acceptable--or welcome.
Twitter is one of those environments where there isn't a public/private binary - much like interacting with people in real life. If I'm sat at a bar talking to friends, there are some situations in which the intervention of a stranger would be appreciated, but it's few and far between. And if they're a salesperson? Forget it.
If I write "I'm finding $productA obnoxious because of $foo; any suggestions for a replacement?", and I get an automated tweet about $productB that shows no signs of actually solving my problem, then yeah, that's spam.
On the other hand, if I got a human reading my tweet (whether they found it automatically or not) and responding with "You might try our $productB; we've solved $foo by doing ...; see $url", then I wouldn't mind seeing that, because they've actually offered me a solution to the exact problem I was complaining about.
Automate and scale it up with ML / NLP to filter and prioritise who to respond to. Hell why not have the bot throw up potential responses.
Just make sure there's a human at the end of the process and really engage.
They are generating actionable sales leads primarily for enterprise business. But I used the trial just as an individual, and that was also sort of fun and more useful than I would've expected!
Spamming people with stock messages is polluting the communal pool, and everyone suffers for it. When you find someone with purchasing intent, TALK to them. Ask questions. You don't need to punch everyone in the face with your sales pitch.
You also miss out on users asking for suggestions who aren't currently using a competitor product (which IMO is a more valuable segment).
A more interesting implementation is one that takes context into account, but that would require some homemade ML work and likely outside of the scope of quick & hacky solutions.
I'd mentioned in the blog post that out of 34 tweets that were added to the spreadsheet, only 6 of them were solid leads. But going through 34 tweets to find those 6 is a lot easier than going through hundreds of them over 8 hours.
The results of the more-filtered-list-tool would be quite interesting, though, as you'd essentially be modeling a set of "ideal leads" and determining how close/far a set of tweets are to those models. Just figuring out an "ideal lead" model for the segments you're targeting would be an interesting intellectual pursuit.
I think I might end up building this...
I have a half-baked contextual analysis implementation which I could probably spin into a high-volume twitter analysis tool. Was doing NLP analysis on unstructured data (like news articles) and extracting topics + extrapolating commonalities between sets. Could be used to pick up topics from tweets and determine if two unrelated tweets are actually talking about the same thing (without necessarily replicating the same syntax).
Definitely worth checking out if you're looking for a more complete Social lead gen solution + no code!
No, they aren't, but not a surprise that automated spam is a big hit on HN.