Wednesday, October 31, 2007

Google Reader update

A while back I posted that I was going to try Google Reader. After using it for a while I highly recommend it. It has cut down on my start time in the morning by seeing what is new in the blogs I'm interested in. I feel like I'm more up to date on what's happening around me. Before I used it, I had just about reached my limit of the number of blogs I could look at. Now if there is one good article that I catches my attention, I will subscribe to the blog to see if more come out. It saves me a lot of time not having to see if something has updated, instead I get told. In fact I can see just now that Patrick updated his blog with a new entry. I love that. I'm not sure how it compares to the other tools out there, but right now it does everything I need so I'm not looking at other tools.

One thing I really don't like is when people use the summary function and I have to click through to their blog. I've completely switched on this issue. I use to prefer being on their page, but now, it just slows me down. There are only a few blogs that are using this function and I rarely go to those blogs. Is almost to the point where I want to unsubscribe. The issue is that the 2-5 sentences is not enough to grab my attention. There could be a great article behind it, but I will not find out because you are interrupting my work flow. So you are using this function, either hit a home run with your summary or you may want to reconsider using the feature.

Product Review

I asked for a product review over at the BoS forums (http://discuss.joelonsoftware.com/default.asp?biz.5.558734.23) and got a lot of responses. According to Google Analytics, I have had 207 unique visitors visit the site and 45 people download the beta (44 really because 1 was me testing it.) Not to bad. The feedback was great as usual from the BoS crowd. They pointed out some minor issues that needed to be fixed (and could be fixed quickly.) I've got quite a few of those done already. I want to thank Joel Rein and Patrick McKenzie for their input. There was a host of others, but they didn't provide site that I could reference.

There were three major issues that came up.

The first one was something I knew about. Patrick pointed out that it really needs to have some sample data. I had been meaning to do this, but it is tedious work. I'm going to try to find a trouble shooting guide on the web and ask if I can use it if I give credit to the author. Related to this is the fact that the app is not really useful until you enter all your data into it. This is very true, but true of a lot of other applications in general. But his point is that it will hurt the conversion rate. I'm not sure what I can do about that.

The second was the limited use of single user app. I knew that going in and I already have the design to make it multi user. But that is a big effort.

The last issue is was from Doug. T who provided some great feedback. He was trying to figure out what the app is trying to be. It is not a issue tracker (it lacks to many of the tools required for it.) Perhaps I overshot with the name a bit. But it was available.

Right now if I had to define Omni Customer Service, I have to define it as a customer service knowledge base tool. Now that might limit the market quite a bit. But right now I don't have the resources to try to make it more than it currently is.

For now I'm staying the course. I'm updating the software to catch the low hanging items. I think have just about all of those. Then I need to flush out the website and take off the beta tag. It will be tight getting it all done before my deadline runs out, but I think I can do it.

Tuesday, October 30, 2007

Omni Customer Support

I decided on a name. Naming products and companies is always difficult, especially when you want to get the matching domain name. For products, this can be avoided a little bit by doing http://companyname/product. But I wanted a independent URL.

I'll post more details after some sleep. I'm a little late on my release. I was hoping to post it earlier and get some feedback from BoS today, but it took a bit longer than I wanted. But in the end I was able to released the beta tonight (this morning actually.) You can find my skeleton website at http://www.omnicustomersupport.com, including a link to download the trial. It is windows only for now, but since I wrote it in java, I will make it available for Mac and Linux soon.

But for now it is feature complete for 1.0 unless I get some have to have requests. Although I did find one bug tonight, well, more of a counting feature that I have to fix. The bug is technical correct, but not to the user. So take a look and give it a try. I hope you like what you see.

I've got a rare meeting for my contracting tomorrow morning, so I will not be able to watch the site and see if I get any downloads, but I know that will be the first thing I check when I get back.

Monday, October 29, 2007

Vacation - BIG problem for mISV

There is a post over at BoS http://discuss.joelonsoftware.com/default.asp?biz.5.557731.24 from a uISV'er about how to deal with vacations. So far I have not had this issue since I'm still doing contracting and do not live and die by my uISV. But I hope that one day that will be the case. But I was excited to read this post because it is validating my customer sevice app. This was one of the use cases I came up with when I thought of turning it into a product.

Lets say you have my app and have all the support issues documented, and you just need someone respond to the customers while you are gone. You hand them the app with your index and let them search the index of issues and respond with the templated responses. Granted you can do this with a word document or a bug tracker, but the key is that search.

My product is built entirely on a search engine. Search is what it does. If you had handed your support person a word document or bug tracker, they will be less likely to find the correct issue. This will lead to more questions/responses with the customer and overall poorer customer service. The reason they did not find the correct problem is because they are doing keyword searching only. It does not take into account fuzzy queries, wild card queries etc. And no weighting is done according to the quality of the match to order the results.

On the progress front, the app is 90% done. I've got one thing I need to change and I think it will be feature complete for version 1.0. I still need to put up a simple web site. I hope to have that done tomorrow so I can post on BoS and get some feedback.

Friday, October 26, 2007

Customer Service App

A lot of progess on my second project the last few days. I created my own icon from a free icon set. I had to modify it a little. But I like the way it came out. But it was only 16x16 and would not scale very well. So I went looking for a larger image. I ended buying one from BigStockPhoto and then modifying it as well. I'm pleased with the outcome. What do you think?
















Thursday, October 25, 2007

Naming Contest

I'm in working like crazy to get my second product completed in two weeks but I don't have a name for it yet. So I thought I'd ask my readers if they have any idea. If I pick your name, you will get a free copy of the product. Let me describe what the product is.

I created the program to help with customer support. I have a strong background in search engines having worked with Lucene a lot. So I wrote a small app that allows me to enter issues/problems, the solution and a template response. When you get a customer service call, you can search your index for the matching or similar issue. Since it is based on a search engine, it ranks the results based on how relevant the results are so that the odds are the one you want will be on top. This is a big advantage of using the search engine over a database query.

I have both the solution and the template because the solution is for internal consumption. It can have things in it that you don't want to tell the customer, possibly a rant about the product, or links to bug reports, etc. The template is designed to be used in the response to the customer.

I also keep track of how many times that issue has come up so you can see if you need to improve the usability, add a feature, add inline help, etc. There will be some basic reporting features to view this data. At first it will be a single user solution, but I plan on making a multi user solution where they share the same index.

I hope to have a few screen shots up soon. Now that you have an idea of what it does, what would you name this product?

Burning the candle at both ends

First things first, the fires in San Diego county are in a much better state then when I last posted. My house has been in the clear since late Monday. With wild fires, that is not 100% true since you can't predict them, but I couldn't see any flames, glow or smoke and I was not in a mandatory evacuation zone. They are still burning but the tide is turning now.

So I went against most of the advice on a recent post on the BoS forum about working on two products at the same time. At that time I didn't think it would be a smart thing to do. Then a few days later I asked if I had a potential second product. I only had a few responses, but they were positive. I took the advice and decided to develop it in java. My goal is to have it wrapped up in two weeks. So far I'm on schedule to hit that deadline, but it has pushed back Web Work Schedules by at least two weeks.

If you are developing a java desktop app, do yourself a favor and go read Patrick McKenzie's blog from the beginning. It has saved me a whole lot of time and has improved the quality of my product.

Monday, October 22, 2007

Extreme Programming (Fires in San Diego)

This brings a whole new definition to the term Extreme Programming. I've been awake since 4:00 this morning since my neighbor called to say they were evacuating. Fires are burning all over San Diego county, including Escondido where I live. We packed up our things and my wife took the kids to her parents house near Irvine, CA, about an hour and 20 minutes north of where we live. I've got the second car packed with the few other things ready to leave if I have to. There have been several homes burned withing a few miles of my house.

I live in a canyon so I don't have very good visibility but I have not seen any flames. A lot of smoke and ash. There has been no mandatory evacuations yet, but I've had the TV on all day monitoring it. At this point it is the best source of information. I can't seem to find anything up to date on the internet. If you know of any sites that have maps of the current evacuation areas and the current fire locations, please let me know. The last one I saw that was any good is over 12 hours old and that is meaningless at this time.

For the most part I feel pretty safe where I'm at. I've got no landscaping (new house) so I have 30-150 feet clearance around my house and no tall trees anywhere on my property. My neighbors and I are setting up a watch cycle to get us through the night. Most of the fire has moved to the south and west of me (downwind), but there is still some of it to the north and east that is of a concern.

I don't plan to get much if any development done in the next few days, but hopefully I'll still have a home office when it is all said and done. My heart goes out to those families that have lost their homes today. I've lived through a house fire when I was a kid, so I know it can all be replaced, but that will not stop the pain and suffering they will be going through.

Sunday, October 21, 2007

Linkerati

Patrick over at MicroISV on a Shoestring has a great post about the Linkerati and his new marketing/SEO experiment at Daily Bingo Cards. There is an amazing amount of details on the how, what, where and why he is doing the new site. For anyone selling to a non technical customers, it is a must read on how to get those precious inbound links. The amazing thing to me is that all of this is for software that makes bingo cards. I think that is about as non technical as you can get. Maybe something for knitting might less technical, but not by much.

He also ties it back to an earlier blog about snowflake queries which I have already mentioned before. I'm running late on getting my own snowflakes up and running. But I should have it in a few days if I don't get side tracked.

So go read the blog entry and learn how Patrick earns over $1000 a month selling Bingo Card Creator to teachers.

Friday, October 19, 2007

Successful Bootstrapping

A big concern with starting my uISV is still paying the bills. At first I was living of savings and was comfortable doing that. If the uISV failed, I was prepared to go back to a regular job. A few months later my old company approached me to do some consulting work for them. I can't stress how important it is to leave your job on good terms. Not only did they approach me for consulting but I knew that I could go back and get my job back at any time. This and several other standing job offers from former coworkers at new companies took a lot of stress away from starting out on my own.

The consulting has been going well, not enough to pay all the bills, but it stretched my savings out by 4-6 months. My current contract is ending and they have told me that they have more work and are willing to give me as much as I want. As a bonus I also gave myself a raise. Now I think I can work less than 30 hours a week consulting, and that leaves me at least 20 hours a week on my uISV and start rebuilding my savings. Since it is consulting work, it will not last forever, so I want to rebuild my savings to have a cushion to fall back on.

One thing I have learned from journey is that I really love working from home. The lack of commute, savings on gas and insurance, less damage to the enviromnent, hot lunches, and very causual dress code are all nice advantages of working at home. But the single best thing is spending more time with my family. My two kids are 3 and 1 1/2 and I get to see them more than ever. Even if my uISV doesn't take off, this time spent with them has been worth it.

So heres to my continued success at bootstrapping my uISV.

Thursday, October 18, 2007

Update your car insurance policy

It is something I should have done a few months ago, but it was pretty far from my mind. I just got my renewal notice (I pay for one full year at a time) and called up and changed my yearly mileage from 15K to 2K. We take my wife's car for trips with the family and I use my car very little now. This is going to save me several hundred dollars a year. So if you are a uISV and working from your house, make sure to update your car insurance policy.

Tuesday, October 16, 2007

The downside of web application

Every browser is slightly different and will do different things. With the slew of javascript libraries and CSS tricks, you can avoid a lot of problems if you stick with the major browsers. But even then you are not safe. I am currently using PopBox, a javascript library for expanding thumbnail images to larger images. It is easy to use and provides a little flash to my site. I thought it was working perfectly, but it turns out, there are problems with the Safari browser, not Beta 3 for windows which I am able to test with. I want to give thanks to Tobin Schwaiger-Hastanan from Cupid's Lab who tested out several versions until I got it working with Safari. I also like to give credit to John Reid, the creator of PopBox. If you need a nice simple java script api to make your screen shots (or any images) pop, check out PopBox. John looked over my page and found out that Safari was having issues with the layout I was using and told me how to fix it. Now I need to get these and a few other changes wrapped up to so I can post it the server.


And then there is IE6. What a fun couple of days this has been. What I thought would be a simple fix has turned into a lot of pain. I've put it on hold while I work on other issues. It almost makes me wish I was making a desktop app instead of a webapp. Oh well. Back to work.

Monday, October 15, 2007

uISV Class of '07

I created a Wiki Page on Business of Software forums (http://discuss.joelonsoftware.com/default.asp?W231) to try to establish a small community of people who have started their own uISV in 07. I figure most of us are going though the same or similar issues and being able to talk with each other, track each others progress, provide feedback, etc. would be useful. So add your name to the page and lets enjoy our journey.

Update:
The wiki page has been updated a few times including:
Simon Huntley from Small Farm Central
Boris Rogge from Sporta Vista
Jon Chase from Send Along
Benjamin Curtis from Catch the Best

Broken Chain

Last month Nick Hebb posted a list of Seven Motivational Techniques for MicroISV’s. I had heard of the Jerry Seinfeld approach of using a calendar and placing an X on each day that you work on the project. I commented that I was going to try this technique and I did. I was going to go buy a wall sized year at a glance calendar at OfficeDepot or the like, but instead I remembered CalendarsThatWork.com and printed out exactly what I needed. A great easy service to use if you need any sort of calendar.

I started on October 1 and had a nice long string of X's going. But then I had to break the string. My family took a road trip to visit both my family and my in laws so I lost 4 days of work. Really 2 full days and a 2 partial days since I typically don't do a lot of work on the weekends. But it was not entirely wasted because I was showing my dad the application and noticed a bug with his browser. When I started his computer, I almost updated his browser (IE6 to IE7) but didn't and I'm glad. For some reason there was some rendering issues on a few of the screens. Its on my list to test more with other browsers besides IE7 and FF, I guess I need to start this sooner rather than later. There was a post on the BoS forums about testing with virtual servers and a few links to services that provided it. I can't find it now, if you remember it, could you post it in the comments. Is it just me or does the search function at BoS terrible? I have a background with search engines (Lucene) and it seems like this should be a slam dunk these days to implement. Someone needs to tell FogCreek about Lucene.NET.

A new week has started and a new chain of X's has begun.

Monday, October 8, 2007

Google Reader

I had a former coworker mention Google Reader to me and I've been meaning to take a look at it. I've finally got a round to it this weekend. My list of blogs and sites I read has increased in the last few months. I started by having a separate folder in FireFox that I would open all tabs and look for new content over my morning caffeine. I know this is not optimal and there are RSS Readers (like Google Reader), but I like the idea of being on the site. Being on the site helps establish in my mind what I'm reader and for what purpose. But this was beginning to take a while, and felt a little wasteful with some blogs that don't update frequently. This morning is my first dry run at using the reader, and so far I like it. The one thing I don't like is it does not show how many comments there are on blog entries. So if I post a comment on a blog, if they have the feature, I have them email when a comment is added, but now I have to check two places for updates and that does not seem optimal. Not sure if there is a better solution. I'm going to keep using Google Reader for the next few weeks to see how I like it.

Google Reader also has a "Share" function which puts the link on your public reader share page, which from the SEO perspective should give a tiny little boost to those links.

Sunday, October 7, 2007

Snowflake Queries

I read Patrick's post about snowflake queries again. If you have not read this post (and many more by on his blog) you are missing out on a ton of great information. I've been thinking about this and wondering how to leverage this on my site. I have an idea. One of the things I have on the list of things to do is create a demo account for people to try out the application without having to input data into it. Some of the feedback I got last week was that for new user, the site is a little empty and they were not quite sure what to do. This is a big issue because if a potential customer takes the trouble of setting up a trial account, and when they get into the application if it is empty, they might simply leave not realizing everything they can do.

So my thought is to create a trial account that is fully populated with data and allow anyone to log into as a manager or employee and see what they can do. Now this opens up some issues. I'll have to turn off some features (email notifications, changing passwords, etc. ) One problem I've seen when people do have trail accounts is that a lot of the time, the data is so bad because people enter random stuff into, and it affects the "realness" of the application. Its hard to image you using scheduling software if the names are foobar and schedules have 18 hour shifts and working 7 days a week. It just doesn't feel right. So I'll need to reset the data every couple of hours or days. Not quite sure which. I'll also need to populate it with realistic sample data. This should increase the number of potential customers who take a more in depth look at my product and the number who will get a chance to see it in action.

And now the snowflake portion. There are few key markets I'm targeting including restaurants. So for each market, I will create a unique URL and page for that market which will be my snowflake query. Its a no brainer now, but before Patrick's post, I probably would created a /demo page and a link for each market. Now I know better.

Saturday, October 6, 2007

Sail Ho!

As a MicroISV, it can feel like you are a small boat in a big ocean. You keep busy trying to sail to your destination and suddenly you spot a spec on the horizon. Sail Ho! Its another ship.

I what to give thanks to Starr at The Startup Lowdown who mentioned my blog and uISV yesterday. While I was on her site I found her reference to Tony Yates at Thinklife. Tony's blog is about his attempt to build a uISV from scratch. As I was reading through his blog trying to figure out what he is doing and where in his journey he is, I found this entry which is about a blog by Ian Landsman at http://www.userscape.com/blog/. A wealth of information. I'm not sure how I missed all the references to Ian in the BoS forums.

I've added these two blogs to my reading list, you might want to do the same.

Friday, October 5, 2007

What I know about SEO

This will be a short post :)
I'm learning as I go. There is a lot of information if you google on SEO or anything like it. You can't go to Digg, Reddit or any other aggregator and not find at least one post on the front page about how to improve your SEO. The trouble is figuring what things are going to return the biggest return for you.

So far I'm relying on two sources of data to help me with starting my SEO. The first is a blog, MicroISV on a Shoestring by Patrick McKenzie. While he has lots of information, I'm starting with a single entry (On-page SEO for Small Companies)and working on that. This gives me a finite list of things that I can work on. I've already started on #1. The changes are not live yet. The source is called WebSite Grader. This gives you a basic run down of things you should be doing and gives you a score on how well you did. I ran it today and got score of 39/100. So now I have a baseline to work with.

The big problem I have is inbound links. I'm not sure how to get them. This blog created my first inbound link! 1 down many to go :)

Thursday, October 4, 2007

In the Middle

This is not the beginning. This is the middle. Eventually I'd like to go back to the beginning and tell the whole story, but for now I'll start here. Here is an almost version 1.0 release of my software, Web Work Schedules. It is an online application for creating employee/staff/labor schedules. I just took the covers off it officially for the first time today when I asked the readers of the Business of Software forum (of Joel on Software) to review it for me. Its amazing what you don't see after looking at software for months on end. Some very big missing pieces on the website. For example, no where did I actually say that it was an online application for creating employee schedules. So I've got some work to do. Some of the issues are pretty minor and I can clean up quickly. Some are are going to take a bit more time to think about. The biggest issue is that at the heart of things I write code. That's what I'm good at. But to make this endeavor succesful, I'm going to need to improve my business, marketing and sales skills.

Luckily there is a lot on the web to learn from. I'm learning a lot from fellow MicroISVers. I read their blogs regularly and have read enough that I have a long task list of things I need to do based on their experiences.