Yesterday I gave the Twitter 10k my full attention – well sort of. I spent 90 minutes on the phone with Matt at Think Michigan Now and an afternoon trip to the gym didn’t give me a full day to work with, but I think that’s probably a good thing for this experiment. After all, all of us divide our time between working in our business and working on our businesses and just like you, I can’t back out of a pre-scheduled appointments or stop replying to emails just to get Twitter followers.
In all, I spent about 3.5 hours yesterday working out my process and getting a handle on the technique. As I mentioned in the first post, I’m trying Joe Mackey‘s suggested method but I know I’m already tweaking it some. For instance, I’m not so hung up on first finding Tweeps with over 10k followers as my base for searches. I can’t possibly max out my followers on any given day, so I’m trying to find interesting people – like those who reply to me or have uniquely named lists. Even lists where I’ve been added as a member because I must have something in common with the other listees.
I’ve also begun to question quantity vs. quality. Quality does take longer to find, but is 10k some magical number that includes a critical mass of quality? I don’t know yet. Maybe this answer will be more apparent as I get further along. Especially if I begin to notice any blips with web traffic or other metrics.
A few observations so far
The follower’s are coming in at a good pace. I’ve decided to start each morning by recording some basic information so it might be easier to spot trends down the road. Obviously, there isn’t a big data set as of yet, but jumping almost 250 followers for 3.5 hours work isn’t a bad first day. On the other hand, you could consider Tweepie’s Flush as 400 wasted clicks. 218 people I didn’t need to add, and 218 people I now have to get rid of, but by getting rid of the spares I think it gives me better numbers. A better ratio of follows/following and it might come into play if I ever have a day where I can max out my follows. So for now, the daily flush stays.
I’m really glad I didn’t have my cell phone linked to Twitter. The amount of Direct messages coming in via Text would have cost me a fortune.
It’s probably a good idea to opt out of Twitter’s New Follower notice while actively working the 10k. I don’t have a limit on number of emails so server space isn’t an issue, but let’s think about this for a minute. The goal is to get 10k people following you. That means 10k emails in your box. My first attempt to wrangle emails was to create a folder and filter to clear all the Follows and Direct message notices out of my inbox. This monring, however, there were a couple hundred that came in and I could see my system struggling to filter my normally heavy volume with the extra 300 emails.
3 weeks isn’t a practical time frame – real time is. If anyone is going to learn from this experiment, I need more than “max out your followers every day and in three weeks you’ll have 10K.” So I’ve added a little bit of geekery to the experiment to help out on that front.
The beauty of a Linux Operating system (which I use exclusively) is that it comes with all sorts of cool software and KTimeTracker was already on my system. I also have the option of working on various virtual desktops. By putting all of the applications I’m using for the experiment on their own desktop and things like blogs, facebook and others elsewhere KTimeTracker can count just the time I spend hunting for followers and none of the time I spend elsewhere. The only special thing I need to do to pause the time tracking is just switch to another desktop – like check my email before I go to lunch.
Now as much as I had hoped for perfectly accurate time tracking with this method, it’s painfully obvious to me that I haven’t used the software enough not to screw things up. This morning, thinking I might want a to take a screen shot and not give away my client list, I deleted the numbers had racked up on this project. Apparently checking the box next to the task isn’t the way you select something for deletion, but having the task highlighted is. There’s also no Undo.
The good news is that I saw the time just before I dumped the data and will be able to tack on an extra 4 hours, 16 minutes to our tally.
The next thing I’m curious about is the morning flush. After I got back from the gym yesterday I did a little more following. In fact I was still following people at 4:30 pm and wondering if anyone would still be adding me after normal business hours. What if I didn’t give them enough time to follow me back? Or better yet, what if I give them a little extra time and then don’t have to re-click them out of my list? My compromise was that I would only flush the tweeps I followed in the first part of the day and give the rest of the list time to marinate over the weekend. I still have that whole business hours thing working against me, but I bet my clicking finger could use the break so don’t see this as a problem.
My final observation for this post – Try to finish mining a Tweepie generated list before you take a break. The system seems to time out after a while probably because of the rapid pace at which data comes in from Twitter. Just before my call with Matt, I made note of what page of the list I was on. When I came back, page 45 wasn’t page 45. In fact I didn’t start getting into new tweeps until page 57. So I sort of resented the fact that I had to search out my end point anyway and I’m still not sure what the method to the madness is. If I come up with any theories, I’ll be sure to share them with you.
As a reminder, my little men are off from school next week and I’ll need to make a quick trip to Michigan for meetings, so if I don’t update you on the process next week, never fear, I will get one out the door for you as soon as possible.
If you have any questions or comments about my Twitter 10k, feel free to leave me a note, below. Otherwise a tweet, Thumbs up or other share using he buttons below would be greatly appreciated. ~Karlie