Posts by Chris Boudreaux

Private sharing defines an emerging class of social applications that let you share content with a limited set of people. While utilities such as Facebook, Twitter and FourSquare tend to encourage you to share with more and more people, emerging applications such as Path intentionally limit the number of people who can see your photos, videos, etc.

What? Doesn’t that limit their network effect — their ability to quickly grow on the backs of their users’ address books, and someday IPO?

Maybe. But it also fills a gap in most of the mainstream social networking applications: privacy. Most of us who use Twitter, or LinkedIn, or Facebook have added so many people that we’re afraid to share our most intimate thoughts or memories. Yes, Facebook offers Groups as a way to control who sees what, but the reality is that most people don’t use Groups very effectively because it’s too much work.

In Path, you are allowed to select only 50 people who can see your photos and videos. While 50 may seem like a large number, most Facebook users have far more than 50 Friends.

Of course, private sharing sites are not completely new. A Small World has long billed itself as, “A private online community designed for those who already have strong connections with one another. By invitation only.” In fact, A Small World has special rules for determining when a member can invite new members. I know folks who have been a member of that service for years and have never been allowed to invite even one new member.

The Future

As these services evolve, some people may divert some portion of their thoughts and opinions into applications which offer greater privacy, such as Path or A Small World. That data may not be visible to search engines or conversation monitoring solutions. On the other hand, some of those services may choose to share the data in limited ways, like Sermo and PatientsLikeMe.

When we help clients empower internal experts to engage in social media, we consistently find that the best way to motivate employees to share knowledge is to simply make it a part of the job. Actually, make it part of doing a good job, and motivate people to do a good job.

Don’t create new incentives for knowledge sharing. Don’t try to pay people or reward them, or give an award to the person who shares the most knowledge.

Why? Two reasons: First, all such incentive systems can be gamed. Second, when you provide incentives which are separate from the incentives of the job, you send the message that knowledge sharing is not part of the job. Finally, these systems just add more bureaucracy, rules and process, without adding much value to the organization. It just makes work more difficult by giving employees more rules to follow, or guidelines to interpret.

Here is what consistently works for our clients:

1. Establish Knowledge Sharing as part of the job. Write it into job descriptions, and include it in annual reviews. If you need your internal experts to enter lessons into a lessons system – make it part of their expectations, and only grant final project QA approval when the lessons are documented. Or, if you require internal experts to own or edit part of a wiki, write it into their job descriptions.
2. Reward people for performing above expectations, just like you would in any part of their job, and include knowledge sharing when you discuss overall job performance.
3. Publicly thank the best contributors. Depending on your corporate culture, you might want this recognition to come from peers, or the person responsible for curating internal knowledge. In any case, such recognition can be included in job appraisals, just like any other type of feedback.

As insurers and their agents follow customers into social media, most insurers are still determining how to best ensure compliance with regulations and business policies. In the last year, insurers started adopting social middleware solutions to empower employees and agents in social media with audit trails and usage boundaries that ensure appropriate compliance. While social middleware solutions do help insurers, employees and agents to comply while engaging in social media, insurers still need to proactively mine the conversations occurring outside of the social middleware to adequately manage compliance risks in social media. And only a listening platform with access to 100% of Twitter data can provide an insurer with reliable, consistent monitoring for compliance purposes.

The Need for Compliance Listening

First, most insurers do not directly control the social media tools their agents use. Many agents and brokers are independent; they are not employees of any insurer. As a result, when an insurer deploys social middleware, many agents will not adopt it right away, and many agents or brokers may never adopt it. That leaves a tremendous volume of agent and broker conversations that the insurer simply cannot control through middleware.

Even if an agent chooses to utilize the social middleware provided by the insurer, the middleware may not integrate all of the social utilities or properties used by the agent. For example, it may not include their blog or agency web site.

Further, insurers who block employee access to social media at work cannot control employee access to social media through personal mobile devices, or on personal time.

And social middleware solutions do not capture conversations between customers, where no employee or agent is directly involved.

In the Conversation Mining™ work that Converseon performs for clients in regulated industries such as insurance, banking, pharmaceuticals and utilities, we frequently discover customers providing each other incorrect information that the brand must work to correct. Of course, a brand can address such incorrect information only if the brand is aware of such information. And developing that awareness requires a social listening platform.

Ultimately, the only way to reliably ensure compliance with regulations and corporate policies within social media is to implement a Conversation Mining™ solution that continually or periodically mines online conversations for the types of content that can indicate a potential violation of policy.

Why the Firehose Matters

If an insurer’s customers, agents or employees utilize Twitter, then the only way to reliably monitor compliance within Twitter is to use a Conversation Mining provider with access to 100% of Twitter data; and that requires an official partnership with Twitter.

As many folks have pointed out, any listening platform without the firehose can only provide insight and access to a small portion of Twitter data. On the other hand, the Twitter firehose gives a Conversation Mining provider access to 100% of the conversations occurring on Twitter.

Anyone without official firehose access must use the public API, and, as Danny Sullivan wrote on SearchEngineLand, “The Twitter API allows partners to conduct searches at Twitter automatically, to bring back data to someone based on those they’re following or tap into Twitter data in other ways. However, the API limits how much data can be requested and does not give access to everything Twitter has stored.”

Therefore, if your listening provider is not an official Twitter firehose partner, you simply will not be able to access most of the conversations occurring in Twitter. Your listening solution will be incapable of mining the majority of tweets passing through Twitter, and, therefore, you will miss the vast majority of potential and actual compliance risks.

And Converseon is the only listening platform in the recent Forrester Wave report with access to the Twitter firehose. If your listening provider asserts that they are giving you 100% of Twitter data, you should ask for documentation.

The Role of People in Compliance Monitoring

According to Twitter’s most recent estimates, approximately 90 million tweets flow through Twitter each day. And the volumes increase all the time.

In 2007, 5,000 tweets were posted each day. By 2008, up to 300,000 tweets were posted each day, and, in 2009, 2.5 million tweets flowed through Twitter every day. In the past year, Twitter volumes grew 1,400% over the previous year. As Twitter COO Dick Costolo recently confirmed, Twitter sees approximately 190 million users per month, and most everyone expects Twitter’s growth to continue.

With 90 million tweets per day (and growing), it is absolutely essential to configure your listening platform to filter out as much irrelevant data as possible, and that requires three things:

1. Keywords framed in boolean queries,
2. Constructed by social data analysts,
3. Who are experienced in working with companies in regulated industries

If you simply assign someone on your team to enter keywords into your listening platform, you will be challenged to filter the data to your satisfaction. Your listening team will become inundated with tweets that match the keywords, but have no relevance for compliance purposes.

Only by using keywords framed in Boolean queries, constructed by analysts experienced in working with regulated industries, are you able to ensure the most relevant data to your compliance management team.

Converseon combines the best of automated and human analysis in our Conversation Miner™ suite, which is utilized by a wide range of leading brands to help map, monitor and understand the vast social media conversation. The Conversation Miner™ listening solutions are designed to be customized to meet specific enterprise environments to provide a central, robust and highly configurable listening platform across multiple languages, regions and organizational use cases to help create listening organizations.

In addition, Converseon was recently recognized as a Leader in a recent independent report by Forrester Research: “The Forrester Wave™: Listening Platforms, Q3 2010” (July 2010). (Free download)

The Complete Solution for Insurers

The chart below illustrates each element of a complete social media compliance management solution for insurers, which includes six important elements:

1. Social middleware that restricts access to features based on policy, archives for compliance, and provides supporting metrics
2. Agents and employees trained and equipped for success in social media
3. Clear policies that are regularly updated
4. Proactive identification of compliance and business risks from online conversations
5. Compelling business insights mined from conversation data
6. Compliance Management staff trained and equipped to enable compliant pursuit of business outcomes

All insurers realize that their customers, employees and agents are engaging in social media. Most insurers want to make it easy for their employees and agents to use social media with success and compliance every day. For insurers with customers on Twitter, Converseon is the only listening platform that can give you comprehensive and reliable compliance listening within Twitter.

Converseon embeds its Twitter firehose access in all of our listening tools and research reports, so our clients can rest assured that they are reliably managing the risks inherent in thousands of employees or agents engaging with customers in the public eye online.

Insurers with an interest in comprehensive, reliable and scalable monitoring for compliance management should contact Converseon at sales@converson.com.

by Ben González and Chris Boudreaux

A lot of folks are still confused about whether they should replace existing TweetMeme buttons with Twitter’s new Tweet button (more info on the Twitter blog), and we recommend the following:

• For content created prior to July 2010, maintain your TweetMeme button.
• For content created during or after July 2010, you can use the Twitter button.
• For blog content, we’re stuck. Blog posts created prior to July need to maintain the Tweetmeme button, but posts created since July 2010 can use the Twitter button. However, the TweetMeme and Twitter plugins do not let you apply the button by post, or based on publish date. You must apply it to all posts or no posts. NOTE: While Twitter has not published a WordPress plugin, a few community members have, for example: here and here.

As of today, the Tweet button from Twitter shows lower tweet counts than existing buttons from TweetMeme. If you replace your TweetMeme buttons, your visitors will see lower tweet counts that on the Twitter button, versus the TweetMeme button.

For example, See this screen shot of Chris Boudreaux’s social media research database on SocialMediaGovernance.com for a quick glance at what the two buttons look like together:

You can see in the image that Chris added the new, light blue Twitter button to the right of his existing green TweetMeme button, and the two display dramatically different counts (210 for TweetMeme and 76 for Twitter).

While TweetMeme has been working with Twitter for months, Twitter began counting “… a couple of weeks before the launch of the Tweet Button. This means links which have been shared on Twitter before July 2010 will not contribute towards the count”, according to Twitter FAQ.

Therefore, if you trade the TweetMeme button for the Twitter button on content that existed prior to July 2010, your visitors will not see the true count of tweets you have earned.

If you use both buttons, you should review the FAQ on Twitter.com to avoid duplicating content. (See: “I want to use multiple Tweet Buttons on my page. Is there anything I should know?”)

For some, the quibbles of style that pop up by cluttering your articles with two Twitter share buttons are irrelevant when the count is king. Others will default to Twitter’s crisp aesthetic.

You should also consider Twitter’s new ability to auto-suggest up to two relevant accounts after a user retweets, as one potential reason for deploying the new Twitter button.

In any case, we are very interested in hearing your experiences and insights which can help to inform decisions regarding which buttons to deploy.

by Chris Boudreaux and Adam Edwards

Update (August 16, 2010): Google PR phoned this afternoon to let us know that this screen shot was an experiment, and that the search results below the ads are organic; they’re just mixed with results from Google Maps. Our point was that the results seemed to exclude traditional organic search results, and that the page appeared dominated by organic results from Google Maps.

Yesterday, for the first time, we saw a standard Google web search results page, without any standard web search results.   That is,  a search for “car rental nyc” returned a Universal Results page showing a map alongside results from paid, local, and books categories (see screen shot below). Not one standard organic result appeared on the first page of results.

(click image to enlarge)

As it expands its Universal Results strategy, Google is testing lots of changes to their main search results page. Most of the significant changes occur in searches that contain a localized search term (such as a city).

First, Google added a “sticky map” in the upper right, which maintained a constant position on the screen, as the user scrolled down the page.  The sticky map appeared to disadvantage paid search results appearing below the third slot, which disappeared behind the map as users scrolled down the page.

In the search results we saw yesterday, there were no organic results until the second page.  When we performed the same search today, we found a single organic search result at the bottom of the first page.

This could be a watershed moment for Google, and it could force local or franchise businesses to use AdWords and Google Places (formerly Local Business Center).

Creating a great web site with strong SEO may no longer take you into the first page of local search results.