Let’s take the public Google Alerts account I created for the food author Michael Pollan as an example. These alerts are sent to a Twitter account, and have generated a good list of followers. The auto-twitter page for this account on AlertRank shows you the different ways you can control which alerts are tweeted.

1. Add a hashtag to the end of each tweet to help people find it with a Twitter search tool. You can add as much text as you want here, and AlertRank will automatically trim the alert to make sure the total tweet is within 140 characters.
2. Use the AlertRank quality score to control the importance of the alerts you tweet. The AlertRank score is based on Google PageRank and a collection of other influence factors. The higher the AlertRank, the more influential the source of the alert is.
3. Select the search terms to include in your tweets. You can collect alerts from up to 1,000 search terms in a single AlertRank account, but you can limit the tweets to just the terms you want.
4. Restrict tweets to those alerts with the right page features. For example, if you want to only tweet alerts that allows comments or trackbacks, you can set these options on. That gives you a set of tweets that are suited for a blog outreach campaign.
5. Depending on the amount of alerts you get in your account, you can restrict the number of tweets to a maximum per hour and per day.

The best part is that all of this tweeting is automatic. Since the accounts on AlertRank are free, you can create multiple accounts, each with their own search terms and levels of auto-tweeting control.

We have a lot of improvements planned for AlertRank, and even though it is free, it will continue to grow in response to user requests. So check it out, and let me know if you want anything else added.

When I created the account, I deliberately made it as clear as possible that this was an automatic process, and not a real human running the account. I used the default avatar, used an account name that explained the purpose, followed nobody from this account, and stated in the bio that the tweets were automatic. I’m sure that if I had adopted a persona by using the name and picture of a real person, the number of followers would have been much greater.

So what is the value of this experiment? It shows that by simply auto-tweeting about a specific subject you can attract followers interested in learning more. 1,570 followers may not seem like much compared to the millions promised by the various pyramid schemes being promoted, but these followers are all self-selected. They want to learn more about this topic. It is the ultimate opt-in lead list. The pyramid schemes are all based on the existence of auto-follow bots on many accounts. If you follow millions of people, a high percentage will auto-follow back, but what is the value of this completely non-qualified list? Bulk growth of Twitter accounts is so early 2009. The real key to Twitter marketing with be opt-in lists like this one.

If you use this technique, you can grow a list of engaged followers, and then add whatever messages you want to the account. It isn’t spam, because everyone on the list asked to be there. Let’s say you are a band that wants to build a list of followers interested in your type of music. Just auto-tweet the appropriate alerts, and then add offers for your CDs at regular intervals. The same idea can be applied to all types of ecommerce.

I told AlertRank to autotweet all the alerts that were about the restaurant, or Boston and Harvard Square. I excluded the alerts about competitors. I also asked for hashtags #harvardsquare and #boston to be added to all the tweets.

The tweets are now appearing, and they are an interesting mix of local news and mentions of the restaurant. It is still early, but I think this will be an effective example of how an automatic Twitter account can create a sense of location-specific buzz for a restaurant.

As an example, a real estate lead may be interested in home listings for a specific area. He could be sent an introductory message, then be placed on an automated e-mail drip campaign where updated home listings are sent to him on a weekly basis.

E-mail drip marketing can be an extremely effective way to brand a business to big ticket leads that may still be in “research mode” for weeks if not months.A lead may be more inclined to do business with a company whose email drip marketing campaign is effective, meaning that the company has sent him helpful information throughout the research period.

All you have to do is create Google Alerts that will generate stories your leads will find useful, select the best one every few days, and email it to your leads as an FYI email. It is a great way of staying in front of them without asking for the order right away. You are a great resource, and your information is always so useful. This will build the trust you need when you do try to close.

You can create a social media version of drip marketing with automatic Twitter delivery of Google Alerts using AlertRank. You can tell AlertRank to post a couple of Google Alerts a day to your Twitter account, and then send an email to your leads telling them you have a free information resource set up for them on Twitter. In between the automated tweets of Google Alerts you can tweet additional messages by hand about specials, discounts, etc. The best part is that Twitter searches for the keywords you are auto-tweeting about will bring in new leads as followers.

Yesterday they tried to change something, and they may have broken one of the most important parts of the Twitter culture. Part of the cocktail party or bar atmosphere of Twitter has always been the overheard conversation. I follow Fred. Fred tweets, “Great post @Sally.” I see that post, and even though I don’t see Sally’s response, I can wonder what that great post actually was, and check out Sally’s Twitter stream to find out. If I agree with Fred, I will follow Sally. This behavior was one of the fundamental gestures in Twitter culture. it allowed Fred to implicitly introduce Sally to all of his followers. Twitter has now turned that off. If I don’t happen to follow Sally, I won’t see Fred’s comment to her.

In fact, they have broken Twitter so badly with this change, that Fred won’t even be able to explicitly point Sally out to me any more. How can he say, “Hey, followers. You need to follow @sally”? Won’t Twitter interpret this as a message to Sally, and block it? Will Fred have to say, “Attention followers. You should follow someone whose Twitter name is Sally”?

I give Twitter 24 hours before they reverse this decision. You can follow the debate on Twitter with a search on #fixreplies.

Let’s take the recent case of Motrin. I know it seems like ancient history to anyone on Twitter, but try to think all the way back to last November when a Motrin ad created a Twitstorm. Now do a Twitter search for Motrin and see if you can find any mention of it. All gone, right? It’s already flowed past and you can’t even scroll that far back in Twitter to ever find it.

Not so fast. Do a Google search for Motrin, as millions of moms are likely to do before giving it to their kids, and what do you find? The first page has a link to the original ad that caused the furor on Twitter. The second page is filled with stories about the reaction of angry moms. That is now Motrin’s permanent record.

According to Google’s Keyword Search Tool, 165,000 people did a search for Motrin last month. Can anyone claim that the same number of people did a Twitter search for Motrin? Even with it’s massive popularity now, are there even 165,000 people in the world who know that Twitter has a search? If your kid was sick and you weren’t sure whether to give her Motrin or Tylenol, what would you do to find out? That’s right, you’d search Google.

But this whole event will surely move off the first few pages of results eventually, won’t it? Not if the makers of Motrin don’t take active measures to repair their reputation on Google. The first page of search results for Comcast still has a link to a YouTube video of a technician who fell asleep on a customer’s couch, and that video was posted 3 years ago!

All you have to do is mine data from Twitter user profiles that you can craft into an easily repeated factoid. The secret to gathering data from Twitter profiles is this simple formula:
intext:”bio * [keyword]” site:twitter.com

Let’s say you work with hospitals to create social media campaigns. You can first determine the number of hospitals on Twitter with the Google search:
intext:”bio * hospital” site:twitter.com

This search currently returns 1,100 results, which can be the foundation of your release. Each of these search results will lead you to a Twitter account where you can gather additional data. You only need a reasonable sample to determine some useful statistic that can be turned into your factoid hook. This allows you to craft whatever message you need to present your case as a marketing firm that can help hospitals improve their social media presence.

If you find that the average follower count for hospitals is low, you can use the press release to announce your white paper on gaining more Twitter followers for hospitals. If you find a few hospitals with a high follower count, you can study their tweets and publish a white paper on the most effective uses of Twitter by hospitals.

Once you establish yourself as a Twitter expert for hospitals, you can create a Google Alert for this same search that will notify you whenever a new Twitter account is created with hospital in the bio. This will provide you with a steady source of leads for your marketing services.

Here are some examples:
353 accounts – intext:”bio * charity” site:twitter.com
385 accounts – intext:”bio * homeless” site:twitter.com
200 accounts – intext:”bio * philanthropy” site:twitter.com
100 accounts – intext:”bio * hunger” site:twitter.com
90 accounts – intext:”bio * poverty” site:twitter.com
65 accounts – intext:”bio * abuse” site:twitter.com
35 accounts – intext:”bio * domestic violence” site:twitter.com

With so much technical and marketing ability on Twitter, what kind of a difference could we make if everyone picked a single account on this list, and tried to help promote it for a day?

Does this mean that Twitter is not useful for watching news, and specifically news about Swine Flu? Not at all. My point isn’t that Twitter doesn’t have a lot of quality information. It is just that informal nature of Twitter seems to bring the idiot in many people. Trying to watch the Twitter stream for “swine flu” gets pretty painful after the 100th repetition of the same stupid joke. What does work well is watching the use of specific words in relation to this event. For example, cancel “swine flu” shows some real agonizing over giving up vacations with families and even honeymoon trips.

In the end both Twitter and Google Alerts have a role, and as long as you can learn to control the results with more precise queries, you can get the best of both tools.