August 21, 2008

Twing: Mining Online Forums & Communities The great promise of the Internet is that we can access the encyclopedic experiences, opinions, insights, lessons, data sets plus the sum total of human wisdom with 2 clicks. Imagine if we could test, research or validate our hunches, opinions, and ideas or solicit insights, recommendations and best bets from people who share our interests, have direct knowledge and experience on topics that matter to us and who roughly reflect our own psycho-demographics? Better yet -- imagine if you could get to this motherlode anonymously without having to actually see or engage these people, hear their whinny voices and funny accents or see their bad haircuts and off-beat fashion choices. How great would that be? This notion embedded in the idea that "everything is a conversation" was probably first articulated in the Cluetrain Manifesto. Now years later everyone from Google to Cuil is struggling with the idea of delivering on this promise and constructing artificially intelligent technology to solve the disambiguation puzzle; essentially the quest to create a thinking machine that can understand and discern nuances of meaning, usage and idioms in common language, much like you and I can. A variation on this theme is the search for the"real, deep and ongoing conversation" that sets the popular, political or personal agenda for consumers, incubates cultural phenomena or influences popular tastes, drives brand marketing and CRM or creates self-perpetuating communities of people with common interests. These communities have been interacting on the Internet on discussion boards and forums since the mid-90s. An estimated 50-60 million people either read or post to this many-to-many communications tool and there is some evidence and some opinion to suggest that they are used and valued as much or more than blogs. Now a start-up called has created a vertical; search engine to find, rank, rate and celebrate them. I spoke with Scott Germaise, the Director of Product Management who explained the raison d'etre as "Twing is built to find and index forums. Nobody else focused on them; so we did." It debuted in beta during March 2008 and is going mainstream and full speed ahead now. Designing their search technology to scan, measure and score forums at the post level, Twing zeros in on conversation threads weighting content, taxonomy, usage, key words and number of interactions to produce faster, better organized access to 10,000 carefully categorized topics of interest and a billion individual posts. Twing is, or aspires to be, to online forums what Technorati is to blogs -- the preferred access and search gateway. The promise is simple -- type in the topic and in a few clicks find a bunch of interactive forums on that topics. Consumers can beeline to their favorite things and hurl themselves into the conversation.Marketers can easily find forums related to their brands and start sniffing the air for trends and monitor and manage their reputations. The Twing guys added a few tools to jazz this process up; they rank the growth of forums so you can pick an active beehive or a backwater. If you register, you can save searches and compare results over time. And using the "Buzz Graphic" feature you can comparatively map a topic, a name or brand against as many as 4 others to quickly see what's hot or what's not. As a dedicated Chowhound voyeur, I can vouch for the wisdom and accurate information that emerges from communities and forums. And who knows? Just maybe Twing will help us get a little closer to accessing the complete font of human wisdom.
Same Old Stuff Sinks Online Catalog Sales Growth The 16th annual DMA State of the Catalog Industry study documents a drop in online sales as a percent of total sales by catalogers and Internet retailers; the first such decline in a decade -- which comes as the industry reports essentially flat performance between 2006 and 2007. This 2% decline in sales attributed to online customers drove the provocative DM News Headline " Where Did Internet Sales Go?" and naturally prompts an immediate debate -- is it real or is it a false decline based on how they counted? Weighing in the side of bad counting is everyone's favorite metric wonk Kevin Hillstrom who says catalog guys under mail online buyers which explains the missing sales. DMA researcher Anna Chernis thinks that the study, the one she ran, might understate online sales brought in by search or social networking, both of which get decent spending from the catalog crowd. Others cite the need for viral tools, blogger outreach and still others think we've maxed out on online buyers. I think the decline is probably real and certainly the result of doing the same old stuff over and over again and expecting something different to happen. If you get as many catalogs and as many online e-mails as I do, you realize that things are tame and templated. Promotions are predictable and pedantic. All the old retail tricks from the weekly rotos have been recycled into digital formats with huge cost savings. There's nothing new, there's nothing different and there's no urgent reason to buy other than price or immediate need. But worse, catalogers and online retailers have not realized the genuine power of the digital media because they haven't invested in the tools to move the needle. Very few make offers based on triggers like price, merchandise or frequency. Hardly any seasonal or lifecycle communications reference my size or the category of merchandise I've previously purchased. Even fewer make offers or communicate on the basis of purchase history data. You'd think after I bought socks 10 times, someone would score me as a sock buyer and make me a special offer! And surprisingly few contact me based on an anniversary of a sign up date, my birthday or based on a product cycle. Surely somebody knows how long those cotton boxers ought to last after repeated washings! And these are just the basic plain vanilla marketing concepts. They are not the creative, inventive, viral, blog-friendly, video or social ideas that all kinds of players are peddling. So rather than gnash their teeth or argue about methodology, if online retailers expect to grow their business, especially in the current climate, they have to dig into the data, truly engage their customers and do something different. Its the most cost-efficient and customer-centric thing to do.

Danny Flamberg

I am a veteran marketing consultant working with leading and emerging brands

The Typepad Team

Recent Comments