Twitter is the poster child for the 'scale first, don't even think about revenue at launch, monetize much, much later' model of startup. In the current climate, ventures like that probably won't get funded. Which is a shame. Twitter is addictive and fun and even occasionally useful. If anybody can pull this business model off, it will be Twitter. It has scale, seem to be moving mainstream and they've even fixed their reliability issues.
But Twitter won't survive if it doesn't find a great revenue model. This matters to all of us.
If Twitter fails to find a revenue model and hits the deadpool, it will have a chilling effect on innovation. That matters to all of us in the innovation economy. Sure, you could live without Twitter, but what about the funding chances for that brilliant idea of yours?
Twitter has top tier VC backing and enough cash to wait till they have scale and the right revenue model. But when they launch it has to be just right. In this climate, Twitter would probably not recover from a Beacon like fiasco. Nor can Twitter do well with a smorgasboard of revenue models, where none really moves the needle and all irritate users just a bit. It has to be one absolutely compelling model that enables Twitter revenue to scale the same way their usage scales.
Adwords is the monetization gold standard. The fact that Google did not invent it is irrelevant. But how many new revenue models like this have we seen in the last 10 (or even 100) years? That's right. This is tough to get right.
But Twitter's challenge is even tougher. Google was not creating an entirely new type of service. It was Yet Another Search Engine, just a better one. They could just use/improve a revenue model developed at a different search engine. Twitter has created an entirely new type of medium, which is what makes it so exciting. So there is no obvious place to borrow a revenue model from.
This requires serious creativity. Yes the words "revenue" and "creativity" were just used in the same sentence, welcome to the new world!
These are the two things that Twitter's Magic Revenue Model has to achieve that Adwords did so brilliantly:
1. Do not irritate/interrupt the user and even occasionally add value to the user.
2. Provide a value proposition that is so compelling that even conservative buyers give it a try.
My horoscope today told me to "pick a controversial topic and ask someone smart what they think about it." Well, I am taking that to extreme and asking lots of smart people what they think.
Wisdom of crowds is "so last cycle". Wisdom of experts is what really works. That is what we have here on ReadWriteWeb, lots of really smart innovation experts. Give us your one best idea for a Twitter Revenue Model and show us how it meets the 2 acid tests defined above.
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/8612
Comments
Subscribe to comments for this post OR Subscribe to comments for all Read/WriteWeb posts
thanks for that info
Posted by: Cristian Romero | October 15, 2008 1:13 PM
How many times I gotta repeat this:
A brnad monitoring API that allows third parties to collect the raw data for product mentions and specific elements of language containing issues of redress.
Make one API and dashboard free and basic, and the paid version feature rich, with connections to BI and data viz.
This model could start making money on day two.
Posted by: Alan Wilensky | October 15, 2008 1:27 PM
API-based push/pull services
Posted by: michaelreuter | October 15, 2008 1:28 PM
This seems kind of obvious to me since the makers of the application, Twitterrific, have found the answer. The "free" version contains ads in Tweets, right along with the people you follow. For the "pro" version the ads are removed.
It's an instant revenue model that can be implemented quickly. I know I'd pay for a pro account if the price is reasonable.
Posted by: Chris | October 15, 2008 1:29 PM
I think every 10th tweet should be a display ad, derived from an understanding of the subject focus of the users following the tweeter.
Posted by: Keith Teare | October 15, 2008 1:29 PM
Well here's my idea: http://www.balsamiq.com/blog/?p=217
As for the acid tests:
- I would love to be marketed to this way. If someone spent a dollar to answer my specific problem, it's likely useful to me. Oh, and I get $50 too!
- as a buyer, $1 is not much for this ultra-targeted (i.e. incredibly successful) way to market: you are basically answering someone's problem with your products minutes after then expressed the need for it.
Posted by: Peldi Guilizzoni | October 15, 2008 1:37 PM
Oops, I meant "I get 50 cents", not 50 dollars. Where's the "edit comment" button? ;)
Posted by: Peldi Guilizzoni | October 15, 2008 1:38 PM
Stick advertising on it of course, isn't that how we monetize everything?
Seriously, though, I think Alan's got a great idea. Companies would pay for that from the customer satisfaction budget.
Posted by: Jason Kolb | October 15, 2008 1:39 PM
The problem is they have ceded so many spaces to others by refusing to capture the obvious revenue.
That being said:
Be a network. That is right be the "micro-blogging backbone" and charge other services for API and Timeline access. That will allow other developers to go after specific vertical markets (realtors, charities, etc) with targeted solutions that add value while not having to build/buy/deploy micro-blogging infrastructure.
If they want the tweets off the public timeline charge a premium.
That being said - given Biz's concerns about making anyone "mad" I'm not sure he will want to irritate developers.
Posted by: Brian Roy | October 15, 2008 1:44 PM
The problem is they have ceded so many spaces to others by refusing to capture the obvious revenue.
That being said:
Be a network. That is right be the "micro-blogging backbone" and charge other services for API and Timeline access. That will allow other developers to go after specific vertical markets (realtors, charities, etc) with targeted solutions that add value while not having to build/buy/deploy micro-blogging infrastructure.
If they want the tweets off the public timeline charge a premium.
That being said - given Biz's concerns about making anyone "mad" I'm not sure he will want to irritate developers.
Posted by: Brian Roy | October 15, 2008 1:44 PM
This would be surprise for many of us that twitter isn't making money yet and How they will monetize one of the best service is still a myth...
I would Suggest to Evan that like your Twitter Concept is new and Amazing, your monetizing idea and idea too should be different and amazing....
Posted by: Ali R Khan | October 15, 2008 1:46 PM
I agree with Alan Wilensky, make a trend analysis tool and sell data to brands / marketing people.
Posted by: pedro | October 15, 2008 1:48 PM
Twitter grew fantastically on the back of conferences and I think that would be a great place to look for a business model, ie. the ad-hoc connection of people who have come together for a common purpose. Basically you charge the groups that will benefit from the connection of people at their events. Starting with conferences you can have a twitter sub-domain where all the people at a particular event can have all their tweets aggregated. You can also sell advertising on that sub-domain for related products and services for all the observers who were not able to attend the event. Conferences are a great place to start but the same thing could be done for concerts, bars, and even companies. So yes they should just copy the Yammer business model as well.
I also second the pro-api idea as well.
Posted by: Karl Long | October 15, 2008 2:01 PM
Great discussion, and like your acid test.
I like @Alan Wilensky "Brand Monitoring API" passes the acid test, but not sure this could be the primary revenue stream.
Wish I had something to contribute... I'll keep thinking!
Posted by: Taylor | October 15, 2008 2:03 PM
Have a global follow for each new Twitter user a la MySpace's Tom. Ensure that people are able to still un-follow this account, aka opt-out.
Charge advertisers, depending on timing of tweet and other super-wicked-cool factors, to send a targeted global tweet out. Heck, even allow a double-sized tweet for extra $$s (more than 140 characters).
They could get creative and do some profiling fields in the sign-up process to give people different automatic follows, and then target the paid tweets that way out.
Acid tests...
1. Do not irritate/interrupt the user and even occasionally add value to the user.
- These would be offers, can meet a certain set of requirements to ensure they are not spam stupid, and don't irritate or interrupt because it is an EASY opt-out (just unfollow).
2. Provide a value proposition that is so compelling that even conservative buyers give it a try.
- It's a tweet, therefore it's easy to execute and could come out cheap to buy.
My 2 cents.
Posted by: Satish | October 15, 2008 2:05 PM
Great comments. I am seeing 3 main threads:
1. "Its the data, stupid". Thanks Alan. That could be it. It sounds "not Web 2.0 orthodoxy" and that alone makes me like it. But how big can this be?
2. "Targeted ads". Maybe. It sounds a bit "Google wannabee". But Twitter is also mobile and real time. Something that got me exactly what I needed exactly at the right time and location aware....
3. Paid service. Sounds like something companies do. But consumers, really? Would kill their growth ramp.
What is interesting is that it probably is a binary choice.
The combos don't make much sense. For example: "free if you will take ads, pay if you want to avoid ads". Lets see if I have the pitch "we can deliver the audience that has no money"?
Posted by: Bernard Lunn
|
October 15, 2008 2:09 PM
Why do you assume that adwords don't irritate people? I find them really, really annoying and intrusive. I hate advertising being forced one me, as do many people I know. I use every piece of ad blocking software available to me and nothing will change that behaviour.
Twitter should go for a low subscription fee a la flickr pro - I would certainly consider paying it, whereas if ads show up I'll stop using the system straight away.
Posted by: xman | October 15, 2008 2:11 PM
-Split the revenues generated by tweets with mobile phone carriers.
-Offer location-centric tweets and offers that people can opt-in to. For example: "Get $5.00 off your pizza at Tony's Pizzaria tonight between 5pm-7pm by mentioning this tweet".
Posted by: Camille Goksever | October 15, 2008 2:21 PM
Twitter is all about communicating just the signal, not the noise. The spirit is in short messages, keeping it simple.
Any monetisation model needs to stick within that and retain the twitter feel we all love.
I suggest then, micro upgrades:
- $0.50/month for Twitter Analytics: who do you talk to the most / how many people visit your twitter page / how many click to your page / blog / how many clicks on all your links.
- $0.50/month Ability to message whole network instantly (would self regulate as if i get sick of messages from you I stop following)
Those are just a couple, but I think each upgrade would have to certain criteria such as:
- Not dilute the experience had by others
- Add value to the user
- Arouse curiosity,
Keep all to $0.50/month or $5/year, these small micro payments would be in align with the twitter community
This isn't a strictly freemium feature, as people already love twitter as it is, moreso an extra feature for certain user types.
Alternatively another model I thought was to use Twitter as a platform for Micro payments (read more here: http://blog.bwagy.com/microfunding/). Where user's can simply tweet a tip to a website.
Thoughts?
Posted by: Ben Young | October 15, 2008 2:21 PM
here's a novel one i thought up. Social Donation System. how it works is, users can log in and make donations, and the level of contribution is reflected by their profile. this is gauged against the average level of contribution of other users making a curve.
so people who don't donate might have a red light by their tweets or something. people that donate a little get a yellow, people who donate an average amount get a green light. people who go above and beyond get a blue star or something special to show that they have made an exceptional effort to support the community.
have it calculate daily or something non-intensive, but make sure as with all donation systems that everything is 100% transparent to users. it could work
Posted by: ryan | October 15, 2008 2:30 PM
Charge the users.
Decide on a value for each message; perhaps per-character. This might be as low as 0.01 cents, or lower. I'd be willing to pay, but that also requires an uptime guarantee from twitter. 3 9's?
Ads might work, might not; I never look at them or click on them.
Posted by: sys | October 15, 2008 2:32 PM
I see revenue in the "Device Updates On/Off". That's kinda evil I know, but as a user it's teh "money shot"...
...I follow 100s of people, but I only have a tiny handful, less than 20, set to hit my phone. Maybe to be lease evil Biz could B2B "Device Updates On/Off" to the carriers, bind it to text messaging packages.
Also, and beta testers of the new Twitter here? Sound off.
Posted by: Todd | October 15, 2008 2:35 PM
Data mining service.
There's all this data about people flowing into twitter that governments and PR agencies would KILL for! Why isn't it being sold to these people?
... google does it ...
Posted by: Swizec | October 15, 2008 2:36 PM
I have to admit my first thought was "isn't that the Twitter team's job?" closely followed by "the world will continue to spin without Twitter." And to some degree, I'm still shaking my head. Twitter is not the center of the universe, so let's not make this into something it's not - if it dies, we'll be fine, things will continue.
That said, there is value in Twitter and a biz model really is nothing more than translating some of that value into money. The question is how to do that without killing the value. To answer THAT, we need to look at who receives the value and how much value they receive.
1) Power Tweeters. These people have thousands of followers. They get value because they can reach all of those people on demand. That's huge from an influence point of view. Value is in the size of the audience... so charge people to have more than, say, 1000 followers AND give them analytics so they can see information about their followers.
2) Companies. Comcast etc get to use Twitter to monitor their brand. Give them tools to better do this and give them some want to integrate this into a CRM system so that issues can be tracked. Perhaps make that transparent and use someone like getsatisfaction to aggregate issues.
2a) Sell anonymized, aggregate data about the network to analysts, companies etc.
3) Give people a way to pay. This might be a refinement to #1 or a different thing, but there's probably an intermediate level between people like me who have a handful of followers and people who have thousands. Maybe it's free up to some number that lets you and your friends and family chat (250 or so?) but $20 per year over that.
Stay away from ads though - ads would kill twitter. As someone says above much of the appeal of twitter is the signal, no noise aspect.
Posted by: rick | October 15, 2008 2:42 PM
Twitter's new topics pages are actually very valuable real-estate that can be easily monetized using Google or Yahoo as an ad partner. The topics pages provide for easy contextualization. I explain this idea in more detail at metaBLOG.
Posted by: Aziz Poonawalla | October 15, 2008 2:58 PM
Beranrd:
Brand monitoring is a large market when applied to the mid-sized intermediaries, not the multinationals that have been the target of many Web based brand monitoring via sentiment analysis companies.
In order to take over this market, the retailers and distributors have to be able to analyze consumer complaints against the retail program incentive systems that they participate in.
Most of the market entrants that are pitching brand monitoring only go for brand owners - they are almost all destined to be out of business in less one year.
you can see the analysis here: http://bizcast.typepad.com/clients/2007/10/open-bid-to-the.html
Posted by: Alan Wilensky | October 15, 2008 3:27 PM
Add a natural language processor that monitors all tweets, extracts what meaning it can, and from time to time responds with information that might be interesting. Obviously the frequency of the responses would have to be proportional to their quality. As the NLP and associated semantic database evolves, users would see the twitter bot responses as hugely helpful instead of invasive.
The key here is to make the advertisements come as a response to user action. Adverts that come at random are spam and should be avoided at all costs.
Examples (Jane is the twitter bot):
Me: dreaming of hawaii...
Jane: @me Cheapest time to visit Hawaii is next month. Flights available from $678: http://tinyurl.com/....
Me: Vista sucks. Why am I still using this thing?
Jane: @me Apple just updated their MacBooks. Check them out at http://tinyurl.com/...
Posted by: Jeremiah Konkle | October 15, 2008 3:32 PM
Twitter needs to partner with different media outlets (Television Networks, Record Labels ect.)
The media will pay Twitter for the amount of people who subscribe to their feed.
Users will subscribe to the feeds to get rewareded with prizes and free gifts from daily competitions.
The media will get to promote their product (a tv show on MTV, or a new movie coming out) through the competitions, the user will get free gifts, and twitter will get paid.
Posted by: Adisa Collis | October 15, 2008 3:43 PM
While several of the ideas above would generate some revenue (targeted ads, freemium / subscription model, selling data / API access), I really doubt that any of these models will create as much value as the service would generate as a loss leader for a company that buys Twitter and bundles it with other products & services.
Twitter is a communication tool. My hunch is that their optimal business model will be similar to online email and IM services. Hotmail, Gmail, AIM, GTalk, and other communication products don't account for much revenue, but they give people a reason to visit Yahoo!, Google, AOL, MSN and other sites on a daily basis. Twitter's best business model is to sell the company to a portal (ex. Yahoo!), a mobile provider (ex. AT&T), or a device manufacturer (ex. Apple). All the other ideas presented here have been applied to online email and IM, but the resulting revenue pales in comparison to the indirect revenue associated with bundling the service as a free loss leader.
Posted by: Joe Lazarus | October 15, 2008 3:52 PM
HAHA... Can I please coin a term?
"Blogsourcing" - An act of asking your blog readers to give ideas because you are so brain dead and tired of blogging.
Usage: Holy cow dude today I blogsourced the problem of monetizing Twitter and I got fucking 150 replies from the suckers.. sweet dude.
Posted by: COP | October 15, 2008 3:56 PM
@Jeremiah Twitter already has this technology. It was part of the IP that the acquired through the purchase of Summize. I imagine that something dealing with semantic meaning will be used in the future by them.
Also, Twitturly is already working on one of the ideas presented on this page. ;-) It is scheduled to come out very soon.
Posted by: Joel Strellner | October 15, 2008 4:07 PM
"If Twitter fails to find a revenue model and hits the deadpool, it will have a chilling effect on innovation."
How about a chilling effect on so-called businesses with no revenue? Or maybe a chilling effect on throwing money at enterprises that don't have any way of returning that investment? That would be a good thing.
Posted by: Walter Underwood | October 15, 2008 4:18 PM
What WOULD NOT work is ads in the Twitter feed itself, but it could work if one of the clients like Twitterific starting displaying ads in the small panel on the RHS of the tweet stream - i.e. very much like adwords.
They could always launch "groups" and charge for them.
Posted by: Philip Wilkinson | October 15, 2008 4:18 PM
People simply aren't going to pay for a service like Twitter. In these economic times twittering is not an entrenched enough social tool that people would rather pay than go without.
The key to monetizing Twitter is IMHO targeted, location-based advertising. The whole service is meant for a mobile device, and apps like Twitterrific on the iPhone exemplify this. How about charging companies to advertise in a geographic area?
Imagine if I'm checking my twitter feed on the go, and also get a "ad-twitter" about a sale in the store next door. This sort of advertisement is worth much more than strict, traditional web ads, as the consumer in this case has *far less* investment in order to act on the ad. The response and return on investment per-eye would be higher than almost any web ad. Especially if these ads are targeted by interest, either through historical keyword searches or whatever. Advertisers also know that their ads are hitting the right people - if you get my ad, you are guaranteed to be within, say, a mile of my stores, and have expressed interest in my sort of products.
This works on desktops also, though less accurately, through networks like Skyhook.
If they can pull this off - get enough advertisers for users to be excited about this feature, and target it well enough to be seamless like AdSense, then this will be a hit. I personally would LOVE to have a service that tells me about things going on in my area, especially if it doesn't bug me about it (i.e. I go fetch this info) and piggy-backs it on data I *already* desire (my friends' twitter feeds).
Posted by: Jerry Wong | October 15, 2008 4:19 PM
twitter, the platform, can eventually be offered to the government and corporations. landing some large contracts there can keep them going.
as for twitter.com, i think they should leverage their own tech and start to hire people as micro-content creators specializing in certain niche markets. within these data streams, adwords can sporadically be injected and each person/data stream can have a companion website.
Basically, just do what other blog networks do such as gawker.
Except the people work for twitter. Twitter can evolve out of its micro-blogging shell and be a brand where micro-blogging is a component of. Likewise, other networks can be started and promoted by twitter, for a fee.
I don't think twitter should try to monetize specifically on the tech platform. They need to look beyond what they have built and just make sure it is a critical piece, but not all that twitter is. That's how I see the twitter evolution and how they can be a successful multi-faceted company.
Sull
Posted by: sull | October 15, 2008 5:14 PM
Well, it's hard to find a profitable revenue model for the web 2.0 conpany.
Posted by: gowers | October 15, 2008 6:25 PM
Ads can be put in as keywords after a tweet.
Posted by: Vishi | October 15, 2008 7:40 PM
Every tweet generates revenue for the cell phone carriers. The demise of Twitter would have an impact on text message revenue for every carrier of any size. Negotiate with the carriers to provide a percentage of text message revenue attributable to Twitter messages.
Twitter could even negotiate bundled packages with the cellphone carriers in which twitter messages are separated from their billing from other text messages. "T-Mobile announces their Twitter package. $3.99 per month for unlimited Twitter messages."
No annoyance impact to the end user. Benefit to carriers through continued growth of text message revenue, and benefit to Twitter through sharing of that text message revenue.
Posted by: Spiffy | October 15, 2008 7:44 PM
I think the money has to come from the thousands of corporate/promotional/media twitter users who want more followers.
While many of these are successful, many more are floundering with only a few followers. Twitter should sell exposure to those streams by inserting some of their tweets into a side panel on the site (or specially marked into the stream). These could be highly targeted by analyzing unique terms in the receiver's tweets. For example, if I wrote about Solar Panels three times this week I might get a preview/recommendation to check out SolarCity's twitter stream.
I think this would be relevant and add value if done correctly. What about you?
Posted by: Nicholas Molnar | October 15, 2008 8:46 PM
You mean when Ev made his routine trip to Japan to steal his Twitter idea he didn't ask them to include a business model. How kuri kuri pa is that anyway. Shame on you Ev. Get back over to Japan. Tell them you must have a business model with the idea you stole.
Posted by: Taki Comento | October 15, 2008 8:53 PM
Here it is:
http://www.chrisknudsen.biz/622/how-twitter-could-become-a-real-business-and-why-they-wont-do-it/
Jack Dorsey, Twitter's CEO, responded in the comments. He has blinders on. Twitter is going to die.
Posted by: Chris | October 15, 2008 9:36 PM
The only redeeming value of such a centralized service as Twitter is the ability for global statistics. That is something that Twitter could exploit.
But beyond that, the business model of a messaging environment belongs to its users. Once we get to the post-Twitter era with a decentralized microblogging ecosystem, the business model will be the same as email : value added for the users and whatever each service providers feels like doing.
Posted by: Jean-Marc Liotier
|
October 15, 2008 10:49 PM
if we do, will they send us some of the revenue?
Posted by: Jeremy Toeman
|
October 15, 2008 10:52 PM
We are already making money from location-based twitter feeds, just like a few people have suggested. We sell local ads that are attached to events updated on twitter, including high school football games. Here is a link to one event (ads excluded on our website, but included when the SMS updates are sent out): http://www.wikimetro.org/Group/Plano_West_Wolves_football
Posted by: Jeff | October 16, 2008 12:10 AM
The question we asked ourselves is not: "how to find a revenue model for twitter" but rather "how to find a revenue model for twitterers".
When blogging became popular, we didn't ask for a revenue model for wordpress, movabletype and the alikes. Bloggers were looking for a revenue model for themselves!
I'm afraid I can't talk more about it right now, but if you're interested, make sure to tell the Magpie your twitter name </SELF-ADVERTISING> :)
http://be-a-magpie.com
Posted by: Jan Schulz-Hofen | October 16, 2008 12:21 AM
Ads are dying because they are too intrusive. And so many people receive their Tweets away from the Twitter site through other clients or RSS that it would be tough to get right anyway.
The API idea has legs, but it's too easy to search Twitter for content as it is. There are third party social media monitoring tools that can already achieve what Alan suggests for Twitter, Facebook and everything else. If Twitter closed the API to them then they'd find another way of doing it.
There is much to be said for the "freemium" model, charging either frequent Tweeters, those with many followers, or through adding some great new features like ability to see who acted on links in Tweets, get more demographic information about followers, more obvious promotion to people you don't know etc.
Whatever, they need to do it quickly before something comes along with these features incorporated.
We have taken the concept of micro-blogging and use it on our site for users to "Tweet" their most urgent business needs. This goes to their contact list, but we also send it to people they don't know who we think have solutions. It's proved to be a valuable way to broker new business introductions.
Ian Hendry
CEO, WeCanDo.BIZ
http://www.wecando.biz
Posted by: Ian Hendry | October 16, 2008 1:11 AM
As long as twitter remains a closed source application I will not be giving them a dime...
I find it not very compelling when a company thinks that they are the smartest guys in the room.
If they did find a revenue model would they share the revenue with the twitter "Community" members that created the content and actions that bring value ?
I think that if twitter were open source more than likely someone in the "Community" would have found a revenue model.
We have integrated Lanconia into the Open Source Social Application that we are building.
We have created a non intrusive revenue model that adds value to users not just to the app.
Posted by: william | October 16, 2008 1:20 AM
I like Twitter. I post a reasonable amount (twitter:parkylondon) and would be prepared to pay for my use of Twitter.
My suggestion (FWIW) is that if you post less than (say) 30 times per month it stays free but over that you pay (say) five bucks per month. If you are a mega-poster that would rise to twenty bucks a month for over - say - 250 posts per month. The model is from Libsyn - it works for them, and me, and I think it would work for Twitter.
If they get people dropping off then so be it. It's a great service. Let's make it pay.
Posted by: Paul Parkinson | October 16, 2008 1:57 AM
How about putting up ads on people's twitter page and sharing the revenue? A simple model would be to allow users to enter their Adsense code, and display a certain percentage of ads with that code. Instant revenue for both Twitter and its users!
Posted by: Martin | October 16, 2008 2:54 AM
What is twitter?
Posted by: usr151 | October 16, 2008 5:34 AM
1 2 Next