Suit Up | SI 694

Profitability

Initially we will want to draw in as many users as possible by creating few barriers to entry. We shouldn’t expect a lot of people to want to opt-in to Suit Up through a paid subscription, or anything similar. Therefore our business structure should try to maintain “free accounts” especially in the short run.

Suit up has the advantage of being very closely related to the clothing e-commerce business, because we are essentially linking to products that can be immediately purchased. It seems only natural to propose that Suit Up would be an “infomediary”, where we can provide recommendations and channels to items on sites such as Amazon and ShopStyle. Suit Up would then receive a cut as an “affiliate” for providing links. The commerce sites benefit especially because we are linking specific items based on a user’s profile, so the probability of interest and purchasing are higher with targeted URLs.

The other alternative that we hope can gain some steam as we get more users (let’s hope!) is to simply put advertising on our site. Since we know demographic information about users and we have rating data, we can track what is important to them and allow retailers and marketers to appropriately target ads. We have space in our interface design for some customized banner ads. Another addition to banner ads is “page dressing”, where the design of the site, usually the background, becomes an ad. Last.fm uses it effectively; whether users like it or not is another question.

In addition to the more passive banner ads, we can also try to embed advertised items within the item lists that the user views. This means we can target advertisements that are more closely related to a Suit Up profile. It is somewhere in between a recommendation and a regular ad. Some similar sites such as This Next integrate ads into the content, but it is very difficult to differentiate. This is a good way to lose the trust of your users. So, if we decide we want to offer an advertised item as a recommendation, we have to clearly mark it as one so that we maintain our integrity as a service.

Other ideas for the future once we turn a substantial population and revenue stream are:

• Offer a subscription for premium content, maybe in order to bookmark more than a maximum amount (freemium model).
• Ban those who use ad blockers–this one is contingent on the fact that we have a solid, dedicated user base already.
• Ask for donations! We would have to explore ways to incentivize people to donate, but this is a plan if the affiliate route is not sustainable.

Sustaining user contributions

We discussed in class that social psychology theories suggest that users are more likely to participate in a community when they feel like they are making a unique contributions, when the share traits in common with the others in the community, and when they are given concrete goals. We have some ideas about possible ways to support these motivators:

• “First!!” Credit. We will be highlighting the person who adds a unique URL/clothing item to our system for the first time. In addition, users will be able to make unique outfits that others can favorite which will be credited to them.
• Neighbors. In addition to generating recommended items for a user, we will be using DirectedEdge to find neighbors for users. They will be able to see that there are other users out there with similar taste in clothing instead of only being exposed to the most popular items on the site overall, which might they might have no interest in whatsoever.
• Setting goals for users. If a user contributes a certain amount to the site, we could offer coupons good for use on purchasing an item that they have bookmarked.

Currently our system primarily encourages users to add and interact with clothing items that they are considering purchasing. Some ideas to motivate them to add clothing that they already own to the site include:

• Neighbor-based outfit advice. We can complement our system’s product recommendations with user generated outfit advice by including a section on the site for users to explicitly solicit advice from users in their neighborhood about an outfit for a particular occasion. This would basically work the same way that Fashism does, except that users who do not meet the current standard profile for online fashion websites would presumably be more willing to participate because they are being judged by people who are like them.
• Virtual closet/fitting room: if we had a really nice interface for this, people might be motivated to add more of the clothes they already own, in order to pair clothes they are considering buying with new clothes or plan out outfits for a trip. (Integration with something like My Virtual Model would be one way to take this a step further, but is probably not realistic given that it requires data that the vast majority of online retailers do not provide. )

How this affects our design and architecture

Our lo-fi prototype made good use of whitespace, providing a clean layout for users.  With the addition of ads, some of this real estate will have to be taken up, but since the ads will be content appropriate and look similar to other features of our site they will not detract from the layout in a substantial way.

Incorporating new features such as ads and affiliate click counting would require a number of small additions to our infrastructure and architecture. SuitUp! currently does not track analytics and we have no way of knowing who is clicking what or how often. ShopStyle pays us based on how many times users using our service click on items hosted by them, yet we do not keep our own records so we cannot verify the accuracy of their payments. To remedy this we have a few options. We could enhance our database can keep track of clicks using referer pages, similar to Fark or Facebook. There are also programs such as Google Analytics that that “gives you rich insights into your website traffic and marketing effectiveness” (http://www.google.com/analytics/). This would allow us to not only verify ShopStyle’s reported referred pages, but we can analyze not only what users like or say they are interested in, but also how serious they are about a product.

Additionally, we’ll need a system to manage our ads. Currently we are storing images on our server, but as SuitUp! expands we will need a more distributed architecture. We will need a similar way of storing advertiser submitted ads, as well as a method of submission. FTP accounts for verified advertisers is one option, and would allow SuitUp! administrators more flexibility and oversight to ad submissions

Down the road, if we are to accept donations, we would most likely look into using a PayPal merchant account. Although it would be possible to incorporate e-commerce features such as accepting credit card payments directly on the site, the cost, complexity, and risk might be too high, especially since our site won’t solely be relying on donations. This would not necessarily change our system architecture, but we should keep track of donors as well as what page they were on when they decided to donate, so that we can thank our donors as well as analyze the type of user a donor is and how we can better incentivize donations.