4 ms·

Ask HN: Best way to send MMS

6 points by c1sc0 18y ago

We're trying to build a website that can send MMS messages to cellphones, worldwide. It does not seem to be a technical problem (plenty of MMS gateways out there, SMTP2MMS providers, etc ...) but a business issue. Basically, the impression I'm getting is that most carriers do not want 'small fish' like us to send MMS messages on their network without approval. I'm disappointed with the state of MMS, makes me think of the locked-in world of WAP 8 years ago. So: has any of you tackled this issue? Is there an easy solution or should we just accept this is going to be messy? Any pointers?

swombat 18y ago
Accept it's messy. As long as big carriers are involved, you're going to see locked down monopolistic behaviours. Basically: - expect it to cost far more than you'd think - don't count on any "big wholesale rebate" - it'll be insignificant even if you hit a respectable scale - don't even bother talking to the carriers until you have a huge volume; as long as you're a small fry, you'll have to go through gateway providers MMS is a still-born technology, imho, in large part because of the telco's involvement. First of all, it doesn't work half the time, secondly, it costs way more than it should, thirdly, it's almost impossible to make a viable business sending MMS's other than by being a gateway.
- gstar 18y ago
  MMS is sort-of solving a 10 year old problem - and it's a 10 or so year old telco answer to the problem of "how do we extract money from customers and lock them in?". It's on the way out, too, being supplanted by email. There's a lot of die-hard MMS users in Europe though, and it's quite common to be sent MMS pictures by people who aren't particularly computer literate and just want to send you a picture. Say, if you're calling up for a car for sale on the other side of the country. Personally, I wouldn't deploy anything new on MMS.
felipe 18y ago
> Basically, the impression I'm getting is that most carriers do not want 'small fish' like us to send MMS messages on their network without approval. EXACTLY! The way we circumvented the system a few years ago (2006) was to get a plain unlimited SMS plan, and then programatically send and receive SMSs through a real phone connected to the server via USB (not sure about MMS, but it is probably similar). This is very slow (12 seconds per message), and most carriers tell you "unlimited" but there is an actual limit in the fine-print. I would recommend this solution only so you can bootstrap it. If you really take off, the carrier will block you immediately, so a business model that can sustain the MMS costs is a must (as swombat pointed out). Feel free to email me if you need more info on this SMS hack.
pmsaue0 18y ago
For 3 years I worked on a startup that built SMS applications. Although we did not work with MMS, I am very familiar with the struggles regarding carriers and just getting set up. Ultimately our startup failed - our costs were just too high, our slice of the pie was too small, and to reach the volume that we would need to become self-sufficient would require a bigger financial investment in advertising than our CEO was able to acquire or willing to make. Anyway.... I read over a long email string that I had with reps from several carriers regarding setting up direct relationships with them. At the time, our company was working with the carrier aggregator (middle-man) mQube - they were then the best, but we wanted to cut them out of the picture. If you don't already know this, carrier aggregators are essentially the gatekeepers to the carriers, and if you want to do ANYTHING substantial, you need to go through an aggregator. The alternative is to have direct-binds with the carriers themselves using lease-lines or special relationships. Through my correspondence, direct-binds are no longer offered, therefore it is pretty much impossible to become an mQube (Verisign became an mQube late in the game only by purchasing mQube). The next problem is that mQube's UI and customer service is pretty horrible, and you have to go through it if you want to get your program brief approved. Getting your program brief approved will most likely be your biggest challenge, and program briefs are like $1000/mo at least. The different carriers have different requirements, etc. You'll probably have to peruse and follow something like this: http://www.mmaglobal.com/bestpractices.pdf http://www.mmaglobal.com/bestpractices.pdf We spent months trying to convince mQube that we weren't going to spam people and that our app was legit. We were a non-subscription based service that supplied information/content only when a user initially requested it. That type of application is in the VAST minority of SMS apps, as most out there are horoscope type subscription services that people have trouble opting out of. In short, you are always going to be playing the carriers' game, and you will probably never win or even break even unless you have BIG BUCKS. In Europe and Japan, if I remember, and it may have changed, there is only one cell provider ie Vodaphone, and it is much easier to get things done. The CTIA has the potential to create some standardization, but I don't see that happening. Best of luck

6 points by c1sc0 18y ago

swombat 18y ago

Accept it's messy. As long as big carriers are involved, you're going to see locked down monopolistic behaviours. Basically: - expect it to cost far more than you'd think - don't count on any "big wholesale rebate" - it'll be insignificant even if you hit a respectable scale - don't even bother talking to the carriers until you have a huge volume; as long as you're a small fry, you'll have to go through gateway providers MMS is a still-born technology, imho, in large part because of the telco's involvement. First of all, it doesn't work half the time, secondly, it costs way more than it should, thirdly, it's almost impossible to make a viable business sending MMS's other than by being a gateway.

gstar 18y ago
MMS is sort-of solving a 10 year old problem - and it's a 10 or so year old telco answer to the problem of "how do we extract money from customers and lock them in?". It's on the way out, too, being supplanted by email. There's a lot of die-hard MMS users in Europe though, and it's quite common to be sent MMS pictures by people who aren't particularly computer literate and just want to send you a picture. Say, if you're calling up for a car for sale on the other side of the country. Personally, I wouldn't deploy anything new on MMS.

felipe 18y ago

> Basically, the impression I'm getting is that most carriers do not want 'small fish' like us to send MMS messages on their network without approval. EXACTLY! The way we circumvented the system a few years ago (2006) was to get a plain unlimited SMS plan, and then programatically send and receive SMSs through a real phone connected to the server via USB (not sure about MMS, but it is probably similar). This is very slow (12 seconds per message), and most carriers tell you "unlimited" but there is an actual limit in the fine-print. I would recommend this solution only so you can bootstrap it. If you really take off, the carrier will block you immediately, so a business model that can sustain the MMS costs is a must (as swombat pointed out). Feel free to email me if you need more info on this SMS hack.

pmsaue0 18y ago

For 3 years I worked on a startup that built SMS applications. Although we did not work with MMS, I am very familiar with the struggles regarding carriers and just getting set up. Ultimately our startup failed - our costs were just too high, our slice of the pie was too small, and to reach the volume that we would need to become self-sufficient would require a bigger financial investment in advertising than our CEO was able to acquire or willing to make. Anyway.... I read over a long email string that I had with reps from several carriers regarding setting up direct relationships with them. At the time, our company was working with the carrier aggregator (middle-man) mQube - they were then the best, but we wanted to cut them out of the picture. If you don't already know this, carrier aggregators are essentially the gatekeepers to the carriers, and if you want to do ANYTHING substantial, you need to go through an aggregator. The alternative is to have direct-binds with the carriers themselves using lease-lines or special relationships. Through my correspondence, direct-binds are no longer offered, therefore it is pretty much impossible to become an mQube (Verisign became an mQube late in the game only by purchasing mQube). The next problem is that mQube's UI and customer service is pretty horrible, and you have to go through it if you want to get your program brief approved. Getting your program brief approved will most likely be your biggest challenge, and program briefs are like $1000/mo at least. The different carriers have different requirements, etc. You'll probably have to peruse and follow something like this: http://www.mmaglobal.com/bestpractices.pdf http://www.mmaglobal.com/bestpractices.pdf We spent months trying to convince mQube that we weren't going to spam people and that our app was legit. We were a non-subscription based service that supplied information/content only when a user initially requested it. That type of application is in the VAST minority of SMS apps, as most out there are horoscope type subscription services that people have trouble opting out of. In short, you are always going to be playing the carriers' game, and you will probably never win or even break even unless you have BIG BUCKS. In Europe and Japan, if I remember, and it may have changed, there is only one cell provider ie Vodaphone, and it is much easier to get things done. The CTIA has the potential to create some standardization, but I don't see that happening. Best of luck