Female 1: Things like most companies evolve by offering something free and then they figure out okay I'm going to up sell a few of my customers. How do you, how do you decide, you know, what to put in, you know, the premium product, how much to charge for it? Do you risk alienation, you know, who are your real customers? Are you just, you're, are your product people thinking of your paid customers as your real customers or the free users because those are probably two very different segments right? And so how do you, what are the best practices about going inaudible? Female 2: There's a lot to be said about that and first if I've got to a mature point where we really learn a lot about what are users are doing on the site. So we have a lot of dashboards and metrics that we watch about. What features are most often used or what are less used and what people like. And then we survey users a lot. We never miss an opportunity to get a user in front of us face to face, send out surveys, follow them on Twitter, look at customer service. Every possible opportunity to find out what users are doing on the site, take advantage of it. And then find out what the most valuable part of what your service is that they would be willing to pay for. On Photobucket a lot of are users are young. Sixty five percent are under the age of 28 and as you learned earlier a lot of those young users don't pay, they don't have access of ways of paying and they don't pay. However with a large user base there is a segment that is willing to pay. You're taking care of their most precious commodity, their photos and videos, they want to feel like they're secure so they will pay for the security of knowing that they're safe and they also want more space, more bandwidth and inaudible ads and all those kind of features so. You kind of test and measure and see what gets people to upgrade and do more of what works and just let go of what doesn't work. But really knowing what your users want is important and it takes time to do that. It can take months and years to really find that out. It's very important to stay in touch with the users and learn from metrics. Man 1: Now if I could pile onto that a little bit. Earlier, the first question you talked about people who are out in the audience who are at early stage in their companies. That's a tremendous, you have a tremendous ability in the early stage to try things out without huge repercussions if they're not successful. Because if you ultimately are successful then any impact, any impact that any of your bad decisions have when your small won't matter in comparison to the scale that you ultimately will be. So the value of the learning you can get from trying experiments greatly exceeds the risk of alienating your, your cust, your future customer base by making mistakes and what those offers are. That runs counter to your intuition at the time where you don't want to run an experiment that you don't think is going to be successful. You don't want to ship a product that you don't think people are going to like. So there's a lot of your inner emotion that's working against what I think is the best business practice there. With India we started out just as an, another example, our first guess was look our product is good enough we can charge for it on day one without a free trial. So we put our pay bar a download, $19.95 you can download the product. We sold, you know, we sold zero copies. And we tried, so we tried that for three days. We thought we had a bug. And, but the bug was in the business model that didn't work. Later on, a little later on, this was now the time that there was an election and a lot of debates between, I think it was Kerry and Bush in the debates, tremendous interest and enthusiasm in the news, debates between these Presidential candidates. We thought well heck we can jump on the back of this, we'll make a Bush avatar and we'll make a Kerry avatar and we'll have them debate and boy are people ever going to love that. So we spent six weeks making the Bush avatar and the Kerry avatar and the little podium and the scene that the two avatars could debate in and then we took out some full page ads for people to see. Nobody ever clicked on the ads. We sold zero copies of these also. So another experiment, absolute failure. And after this experiment we were realizing, you know, we didn't even have to make the avatars, we could have just put up an ad first and if nobody clicked on the ad we saved all that time and if anybody, I'd say three people clicked on the ad. Well that's not a huge success. We could just email those people personally and apologize. Man 2: We, we, we did that. We actually used google adsense, we put fake ads out there for fake products and we saw which ones people clicked on and we had a page that was kind of like oh, sorry there's an error. It was basically an electronic apology. Woman 3: That's one way of market, doing innovative market research. That's great. Woman 2: Yeah at Photobucket we found once we got to material level our users were very unhappy if we experimented too much on them and our premium users were definitely the most vocal. They paid, they've got to pay, they tell us what they think all the time so we started a completely different site for experimenting called Tinypic which we never advertised, never promoted, never did anything but it was our experimental thing. And if worked on Tinypic with a small number of users we'd move it up to Photobucket. So you really needed to experiment but we got to mature point where we really couldn't really experiment as much as we would like so we made a brand new site just for experimentation. So it's really good to keep testing and know what your users are telling you to do. Man 2: In terms of what to put in the free versus paid model, I love to keep lists and my favorite list I've been keeping for the last two years is features that our, our users write to us, they're very vocal and features that people ask for that where they also say they'd be willing to pay for that. That's my favorite list and I couldn't wait to finally bundle a lot of that stuff up into a premium service. But, you know, the flip side of that is I think you just have to try stuff before you really know. That's my point earlier about it just imparts a lot of truth in your business when you start charging money because half those things people said that they would pay us for, they don't. You know nobody really wants to pay and there's a lot of false positives there so you just have to get out there, do it, see what they react to. What I'd say has been much more valuable testing for us is putting kind of fake product line ups out there with fake price points and doing a lot of AB testing and seeing where people really willing to go to the next step in the billing flow and that gives you a much, you know, more accurate prediction of what's of value to them. And we did a lot of that testing and that helped fine tune our thinking about what would the premium product be that we launched with.
==== Transcribed by Automatic Sync Technologies ====
