Discuss Home · Bootstrapped Podcast · Scribbleton Personal Wiki · HelpSpot Customer Service Software · Thermostat NPS

Hi, I'm Terrence and I'm a marketing novice


#1

Hi,

I’m a single founder and creator of http://gitsense.com and more importantly, I’m a marketing novice. To say the marketing aspect of my bootstrapping adventure has been a humbling experience would be an understatement. The biggest challenge I’ve found so far has been trying to convince people that my product is not like product X. It’s one thing to try my product and then say it’s like X, but to just say it is, without even trying, has been a baffling experience for me. Copywriting is hard and I definitely need to work on that.

The good new is I’m pretty technical. I have 15 years of software development experience and my area of expertise is building productivity solutions for large companies (1000+ developers). The core technology for my current product is actually based on my previous research on how to reduce information overload and speed discovery through adaptable tools.

Basically every tool in GitSense is designed to be viewed through a window and through the window you can only see what the window maker wants you too. For example, you can create a Release 1.0 window and have it only show links, commits, diffs, discussions, continuous builds, etc. that is relevant to Release 1.0 If that is clear as mud, I have an interactive GitSense Navigation Tutorial that better demonstrates how the GitSense tools are able to display different types of information, depending on the window that they are view through.

Well that’s me, my product and my current marketing hurdles.

Cheers,
Terrence


#2

Hi,

It looks really impressive and the presentation is very good.

I’m not even on rung 1 of the bootstrapping ladder yet so am no expert but a few immediate thoughts jump out.

  • Who exactly are you marketing to?
  • I use Git daily, whilst I can see it looks quite cool it’s not clear to me what pain it’s removing?
  • How would I integrate it into my current workflow? Is it designed to be standalone or does it slot into other tools? Will it slot into Gitlab, say?
  • There’s no pricing indication, is this on purpose?

As I say it’s looks really interesting and polished but I’m left wondering what where it’d add value. Maybe you could start by adding some dumbed-down “Problem X: here’s the solution” type content.

Good luck.


#3

@Terrence can correct me, but this looks a lot like Atlassians SourceTree. So it does solve a pain point, at least for people like me who hate command lines (even though the day job is in Linux), and prefer GUI frontends.

I tried SourceTree, but found it too confusing. If there was an easy to use alternative to it, I’d happily use it. At the moment, I use Github’s GUI client. Its very basic, but that’s what I love about it. I never have to check the manual.

My problem is that I use Windows at home, and there is no Windows version.


#4

Thanks for the feedback @bealers The target has always been enterprise but I use it personally for my own development. The three things that I personally use it for and has turned into crack for me is the revisions tree, the context aware searches and tree diffs.

The revisions tree basically takes all the commits in my search and turns it into a navigable tree. I tend to iterate at a crazy pace and I commit very often for safety sake. So when something stupids happen, I’ll normally just tell the tool to show me all the commits that were made in the last 48 hours and convert that into a tree. With the tree I can see what files were modified and how often. The tree also lets me quickly point and click to select revisions to diff.

The context aware search lets me do a commits, diffs, text, and code search against what I’m looking at. For example, if I’m looking at all the commits from the last 48 hours and do a search, it’ll just search those commits and changes. If I’m looking at a diff between two trees, it will just search the two trees. This makes it very easy for me confirm if something exists at either points in time. This feature is actually the result of my work in enterprise where the code base is generally massive, with a lot of contributors.

Being able search across more than one point at a time, makes debugging easier I find. And it is important to note these points can come from different repositories. In enterprise, a product is usually made up of multiple branches from different repositories and because my tool is repository agnostic unlike GitHub and other browsers, they can easily construct a logical view with different branches from different repositories and search across them.

And the diffs tree is basically how I do code reviews. And unlike GitHub and other solutions, it is designed to review massive changes which happens from time to time in enterprise. It also supports reviewing diffs at the directory level so you can easily sign off all the changes below a certain directory.

It’s standalone in that I don’t care if you use GitLab, GitHub, BitBucket, etc. Its job is to just break apart what is in your repositories and present an easy way for you to interact with your changes. I think of it as Business Intelligence for Git and software development.

For example I generate code churn metrics (lines added, changed and deleted) and outside of Microsoft, I believe I’m the only other commercial product that does this. And I’m pretty sure I’m the only one that offers SLoC code churn metrics, which ignores comments/blank. For example, if you make a few hundred files changes and update the copyright date from say 2013 to 2014 in your codes comment line, my SLoC Churn will show 0. This type of metrics is very important for release primes and senior architects.

I also have what I call Smart Attributes, which are basically programmable context aware metadata. You attach these to diffs, commits, GitHub pull requests, etc. and you can program them in JavaScript to react to what they are attached to.

For example, you can program a Smart Attribute to raise a pop up box that say DO NOT USE THIS COMMIT if it has failed a build and if user Bob is looking at it. I have bunch of examples in my GitSense Navigation Tutorial that contains Smart Attributes and you can view their source and run them against test cases.

One of the things that I would like to do is create a market place for people to sell their Smart Attributes.

I definitely want to make it very affordable for startups and individuals but price accordingly for enterprise. I’m think about doing what Atlassian does, where I make it dirt cheap to use if there are less than 10 users. The catch with my pricing model is I have to consider the number of unique blobs. I charge by blobs because half my product is an indexer and the more blobs you have, the more things can go wrong.

I believe this is how most Business Analyst software price as well. They charge by how big your database is. The bigger the database, the more complex the number crunching.

Yeah I really need to but my resources is being stretched to the limit. I really want to create learning channels that says if you are developer, go here. If you are a team leader or manager, go here. If you are graphics artist go here and so forth.

I’ve been thinking bootstrapping may not be the right route for me. My product is similar to Atlassians https://www.atlassian.com/software/fisheye/overview and I think they have 30+ people working on it. I’m obviously biased, but I believe my product is far superior in usability and features.

It might just make sense to apply to YC or spend my time looking for VC money. Also GitSense is just a by product of my research, as the real technology lies with my Information Flow Control System. With this system I can see better bug trackers, better code review tools, etc. In fact it works so well right now that I can integrate better with Atlassian’s product than Fisheye, which is produced by Atlassian, can.

Thanks again for the feeback.


#5

@shantnu

No it’s more like https://www.atlassian.com/software/fisheye/overview

Git GUI’s and Git Hosting is a very saturated market. It won’t be long before a decent free Git GUI comes along. And GitLab and other free alternatives is really going to eat away at Git Host providers as well. My focus is with analysis.

I just want to make it insanely easy for individuals, teams and organizations to be able to answer questions regarding their codes history. Be it software metrics or debugging why a build failed. My focus is enterprise where you have hundreds of developers and potentially millions of lines of code.

Being able to track what is going on at this scale is not trivial and large companies easily spend millions a year in developing inferior in house solutions to help them manage. Most solutions that have been built would revolve around ClearCase and Perforce but most want to get off ClearCase and Perforce because they are expensive. A ClearCase license costs about $3,000/year/user and large companies normally have 1,000+ licenses.

What I’m trying to do with GitSense is provide a viable option for companies to switch to Git since I can give them the analyst tools that they would need.


#6

I was faced with the same decision. I came close to getting some angel investment, and thankfully realised that it wasn’t really how I wanted to work. This might not apply to you, you sound like you really understand the problem; but if I had scaled up at that point, it would have been FUBAR. Complex solutions to complex problems take careful, slow, and considered development; that’s hard to do if you’ve got 30 developers and a VC waiting for results, yesterday.

If this is where your focus is, forget the website. It’s fine as it is. What’s more: enterprises with hundreds of developers most likely won’t find your website, anyway. (Even after you talk to them, they probably won’t look, because they want you to hand deliver the info to them.)

Go knock on some doors of people managing 100 person teams. Many will be interested but want a finished, polished solution. That’s fine, thank them for their time and file the contact away for later. You want to find that one guy who has been waiting for this tool, dreaming of making it themselves on the side. Help him fulfill his dreams (and charge a lot of money.)

And forget the $10/month plan for startups. You don’t have the time to address multiple markets.


#7

@danielstudds Thanks for the feedback.

The good news is I would need no where near that many people. My technology is at a point where I just need 5 more developers to really act on my vision. I’ve worked for companies with 1000+ developers and what I’ve found is, after a certain point you just get diminishing returns. If anything, the VC money would be used to hire sales people and technical writers.

The reason why I mentioned Atlassian is they actually have an amazing business model for selling to enterprise. Enterprise sales is traditionally top down, where sales guys would shmooze execs and that would get the ball rolling. IBM and Microsoft and other big brands normally go this route. Atlassian on the other hand, takes a bottom up approach.

The bottom up approach basically involves targeting disgruntled employees/departments who can’t stand what they are currently using and are willing to pay for a better alternative. And this is how Atlassian’s business model becomes relevant. In fact, this is how Jira infiltrated one of my previous employers which had 30,000+ employees. Enterprise solutions usually meet the needs of the purchaser, but not the day to day user. And this is how you create a disgruntled employee.

I know what the Git landscape looks like and like I’ve mentioned earlier to bealers, I’m a one man dev team and my product has turned into crack for me. I know this tool can have serious impact to individual developers, which is why I want to make it accessible to them. The only thing I’m worried about is support.

Seeking VC money or applying to YC for me would purely be about advertising right now. I have a very good track record with designing productivity solutions and I know if my product is just allowed to soak for a while, it will turn into crack for others.

Thanks again for the feedback,
Terrence


#8

Marketing, the Achilles heel of many a technically minded person, myself included. My approach has been read, learn, try, test, repeat. It’s like development only with people and emotions instead of code and logic.


#9

This is easily comment of the week. :smile:

Yes, it’s easy to become awed by huge companies with dozens of engineers. But the old saying “Too many cooks spoil the broth” comes to mind here. How many of those 30 engineers are wasting their time trying to fulfill the VC’s pet project, how many fighting their boss who wants to shoehorn in a dozen features without testing?

The more people you have, the more egos, politics and in fighting. While that doesn’t mean the small guys/gals always win, it does mean the advantages of being big are overrated.


#10

Ha, you say you’re a novice marketer, but your enthusiasm for your product is infectious. If you’re passionate about small dev shops, then maybe I’m wrong and you should go for that market instead. Just share your passion and it’ll take care of it self. You’ve got the makings of two great blog posts in your responses to this thread, for starters.

Can’t wait for the mac version!


#11

@danielstudds

Well the fact that I had to write so much is pretty much proof I really don’t know that much about marketing.

I was hoping to have simple installer in place by now but it’s clear I need to work on the website. Your questions along with others might have been the breakthrough I needed. I think this next iteration on my website will finally be able to address that simple question of how can it can help me.

Thanks again for the feeback


#12

Pick a group of people to target. Probably one that’s not locked into github. Say, php developers. Then go to every PHP conference you can find, hand out t-shirts. Do talks. Etc.

Developers are pretty easy to market to if you have a really good product, since they tend to clump together and they like to share info about the latest and greatest tools. And it’s surprisingly easy to get brand-name customers since small technical teams often make their own tooling decisions.

I imagine it will be really hard to sell to people who already use git. I do. It works. Frankly, it would take a disaster to make me switch away from github. I personally don’t really care how much better my git experience could be. I just use the command line and it works.

But there are probably still a lot of groups of devs who don’t use git and who would benefit from it, and from your product.


#13

@starr Thanks for the feedback and I really do appreciate it.

Actually GitSense is designed to be a complimentary asset to GitHub. The more people using it, the better chance I have at selling.

It is quite clear that I’m not advertising what my product does and how it fits in with Git and GitHub. I’m updating my landing page right now and this is what it looks like right now:

http://screenshots.gitsense.com/new-landing.html
http://screenshots.gitsense.com/new-landing-2.html

In the second link, you can see an example of how you can use GitSense to work smarter with GitHub. For individuals and very small companies, being able to quickly gauge risk may not be valuable, but in enterprise, identifying and reducing risk is extremely important.

Thanks again for the feeback,
Terrence


#14

@bealers, @shantnu, @danielstudds, @starr

Guys,

I hope I’m not overstepping any boundaries here, but since you were gracious enough to comment on my post. I was wondering if you can look over my updated landing page at http://gitsense.com. I’m hoping it is more obvious as to how my product would fit and/or not fit into your work flow.

Thanks,
Terrence


#15

Sorry for the lack of reply, @Terrence, got busy. Seems to be a problem with the site, I’m just getting a blank page?


#16

@Terrence it looks like the website is down this morning


#17

@danielstudds @paulyoder

Guys,

Thanks for looking but I’ve replaced my website with a blank page for now. Provisional patents are being worked on for my technology, so I’m keeping things a low profile for now.

Thanks,
Terrence


#18

Bold! It’s something that I’ve thought about, but the time & expense doesn’t seem worthwhile.

What’s your reasoning behind getting a patent?

Also: What jurisdiction are you getting the provisional patent in? Are you seeking legal advice? Writing the patent application yourself, or getting it drafted?


#19

@danielstudds

Getting a provisional patent costs about $1,500 with a good lawyer and it’s good for two years I believe. You should also know that you only have one year to file a provisional patent once you make your product public.

My reasoning for the patent is GitSense is not only a product, but it’s a proof of concept for my information flow control system, which is what I want to patent. My long term goal for my company has always been to provide a suite of affordable enterprise tools that covers the software development life-cycle. Pretty much do what Atlassian is doing and with my information flow control system, I’ll be able to do the following that they and others can’t, which is:

  • reduce information pollution
  • speed discovery
  • solve the problem of “If you try to please everybody, you’ll please nobody”

By patenting my information flow control system, I’ll be able to ensure Atlassian, GitHub, etc. won’t be able to emulate what my product does. I guess it was good thing that my marketing sucks.

If you do have something truly novel, you should consider patenting it and getting a provisional patent is pretty cheap. Protecting your patent will be difficult though, but you’ll make yourself a little more inviting to be bought by somebody who will have the resources to protect your patents.


#20

I like this approach. I thought I’d share a few lessons I’ve learned along the way in my experimenting:

  1. Start with your audience in mind: what are they hiring your product to do? What pain does it solve for them?
  2. Make that primary pain point your headline for your landing page.
  3. Focus your headline and landing page on your audience, not on your company / tool.
  4. Start blogging: focus on the pain(s) that your audience has.
  5. Your primary focus should be building a mailing list (launch list). Your mailing list will help you engage with potential customers, offer value, and eventually, convert them to paying users.

Hope that’s helpful!