Running a Bootstrapped SaaS - And Finally Coming out of Heads-Down Mode!

Hey everyone!

I’ve been a big fan of these discussions for years but never took the plunge to actually join in the fun until just now.

The past few years have been a pretty crazy and wild ride, going from being a solutions architect at a Fortune 500 to building and launching a handful of SaaS applications. I was in pretty heavy heads-down mode during that time and now that I can get a breather, I wanted to come up for air, be a little more social, and finally get around to doing things I’ve been meaning to (like joining these discussions :slight_smile:).

I’m spending the majority of my time now on Tamboo, which is a SaaS app that lets you record and watch what people do on your website so that you can see what might be preventing you from getting more signups or sales based on visitor behavior.

I’m more than glad to share my experiences and knowledge with anyone and hopefully help to answer any questions any of you might have about getting a bootstrapped SaaS off the ground or how to go about escaping the 9-5.

Thanks again - I’m super excited to finally get in on these conversations!



Landing page looks good.
How is tambo different from crazy egg, etc.?

How does it handle.heat maps when different users have different screen sizes and thus layouts?

Fyi, one issue ive had using these services is that I dont have time to watch enough of them to get statistically valid data. So watching them at high speeds os helpful.

I just had a hard time getting actionable data out of it.

Your home page made a good impression on me. It quickly and clearly told me what the site does, how I benefit, and how to get started.

I think I’ll have to try this very soon on my own website.

Interesting thing.

In my web application the performance impact is very strong tho. On page load I’ve got a “unresponsive script” warning from FF, then each action was followed by noticeable pause.

If you want to see my session for reference, here it is:

Also, I’m afraid the number of session could be more than a mere human can handle. I wonder if it is possible to have some sort of navigation map (lines or coloured areas going from point of action to point of action) to see the overall ways how people work, and only then pick those that are unusual for close review?

1 Like

Hey Clay, thanks for the feedback!

I’m really interested in what you said about getting actionable data out of a service like Tamboo because that’s something I have been working towards recently - putting together actionable analytics that you can then drill into to get more detail (watching videos, looking at heatmaps, etc.).

Out of curiosity, what kind of data were you hoping to get out of these kinds of services, and for what purpose? Would really appreciate your thoughts.

Hey Vlad, thanks for giving it a whirl with your app! I’m going to dig in here a little bit to see if I can’t figure out why you’re seeing that with your app on Firefox.

Thanks Steve! I appreciate that!

I gave it a shot. Yes, it is tracking how users use my home page…but I’m left thinking, “What now?”

I feel like I need a short guide or two, directly at hand when I’m viewing sessions to tell me what problems to look for. Do you plan to add any automatic analysis?

Could you add IP address to the sessions? This would help me quickly spot the sessions that were internal users. Failing that, you could specify city + country.

At risk of sounding a bit harsh, and negative, I wouldn’t pay money for this YET, because I have no idea how I can use it to get more sales. But I feel that it certainly is heading towards being a valuable tool.

Hey Steve - Thanks, I appreciate the feedback! I don’t take it as negative or harsh, no worries. If you can’t see how to use it to get more sales, that’s on me, not on you.

I honestly wasn’t expecting to have such a big response as I’ve been getting in this thread, but it’s been very useful.

Let me throw a little more information out there just for the sake of the conversation:

One of my goals with Tamboo is to offer - as Clay put it - actionable data that can do automatic analysis on user behaviors to help people optimize their sites for different goals such as conversions and sales. Those features are in the works and coming shortly.

To date, most of my customers are people who “just want to see what’s going on”. They’ve got traffic coming to their site, but they’re not getting a significant number of conversions/sales and they’re like “WTF?”. The video playback helps them to understand what’s going on instead of just getting frustrated, and helps them see where they might be able to make some changes in messaging, UX, etc. I’ve also seen Tamboo used for non-sales purposes, like people launching an MVP and wanting to be able to see how customers interact with it.

Anyways, long story short, the customers I have today have provided me with (troves of) data to use towards building out the next phase, which are the analytics. Based on what I’m hearing from you and Clay, I’m thinking those features are the ones that you’re really looking for, and although it sucks I don’t have those in place today for people like you, it’s encouraging to hear that my thought process is on par with yours and that those features could help a whole new group of people.

I kind of having an idea - knowing how my users use specific UIs would help me to understand what features take further improvement/whatnot and what is not used and can be moved into less prominent place.

But I cannot think of a practical way to figure it out without watching hundreds of hours of users videos. That I do not have a time for.

So there is a potential - if that bottleneck somehow can be optimized.

Thanks Vlad - great feedback! It seems that there is a huge need to “summarize” this information to make it more useful “at a glance”. I’ll definitely be upping my efforts on getting the analytics features out the door here.

Some feedback to stick in your todo-maybe pile… :slight_smile:

  • When I’m viewing a session, I can’t easily tell if the page views are listed oldest first, or newest first
  • Showing time durations in seconds with 3 decimal points is kind of unnecessary. 0 or 1 decimal points is all I need.
  • Landing page is good, but I’d also like to know exit page. This would help me pick sessions to review.
  • When I click on a session, I’d like the first page view to be already showing, ready to display.
  • The session paging shows only 15 sessions per page. I’d like to be able to see far more than that, so that I can quickly scan to find interesting sessions to view.
  • I think you need some filters so that I can easily find sessions that were less than a certain duration, more than a certain duration, started with a certain landing page, came via google, etc.

How much are you thinking of charging? And why are you not already charging - or at least having a “BUY NOW” button? :slight_smile:

Hey Steve!

Great info there, much appreciated! I’ve got a few of those items on my list already, so it’s great to know I’m on the right track there. Will be adding yours as well :slight_smile:

I am currently charging people for the service, but I’ve been playing with different approaches to get more people in the trial door and ultimately convert to paid. Playing with funnels and all. I’m actually running an experiment right now on the landing page to see if not publishing pricing has a positive or negative affect on the signup rate for the free trial and on the conversion rate for trial to paid. I know that might sound odd, but that’s why I thought to test it. You never know until you try.

Regardless of that experiment, after the free trial, users get prompted to select a plan in order to continue using the service. Plans are monthly starting at $29/mo going up to $149/mo based on usage. So it’s like meter-based pricing with caps on how much data will be captured and retained.

I’ll let you all know how the experiments pan out if that’s something you’re interested in.

Thanks again - this feedback is awesome!

If I knew what the actions of the actionable data was, I wouldn’t need it :slight_smile:
Seriously, I just don’t know what to expect.

One big concern is that different users have different resolution screens, so a click on a button might be in different places on different screens, so the heatmaps would seem to be undependable. And, in fact, I’ve seen heat maps where they show a bunch of click in white space maybe NEAR a button.

So… I see all those clicks in empty white space and wonder : are my users spastic or is the heatmap wrong?

(or better, case studies with A/B outcomes data)
I think a HUGE selling point would be to show examples of what a user saw and what actionable insight they got from it.

You could even make this more engaging* by showing:

  1. The web page
  2. Ask "what is ONE change could YOU make to greatly improve [Action] by [%] ?
    Bob made one change (took 2 minutes) and increased (action) by [%]
  3. Then show them Tamboo’s info
  4. Then ask again.
  5. Show them
  • There is good research (see Pre-suasion by Cialdini) that if you present a mystery to the audience you get FULL attention. They’ll be dying to know the cause of the mystery.

Yeah, there are actually a lot of different screen resolutions out in the wild, especially with mobile and tablet users. Responsive design adds a whole other layer of complexity to the mix because now you’re talking different layouts (potentially at every unique resolution) you have to account for as well. It’s definitely not for the faint of heart.

I’m on board with you that a lot of heat map products out there lack accuracy. That’s one of the things that I’m very focused on getting right. So far from what I’ve seen, the screen captures and playbacks in Tamboo are much more accurate than the majority of the heat map products out there. I’m working to get them pixel-perfect, but as you’ve alluded to, there are a lot of complex factors involved. I’m close and getting closer on that goal. That’s one of the reasons I haven’t published the aggregate heat map analytics yet - I want to make sure I’ve got it as close to perfect as possible before throwing my hat in that arena.

One thing that I’ve noticed in watching a lot of these recordings is that there are some users that are spastic mouse clickers. It’s kind of comical, actually. But if you take something like Google Analytics’ really poor heat map attempt (which is buried in their menus), you’ll see they don’t track actual mouse clicks - they just track the link that gets clicked and try to make it look like it was a mouse click. So if you have the same link in the header and footer, you’ll see that they’ll say the user clicked both links in their stats.

I like your recommendations a lot about showing examples with tangible outcomes. This is something I’ve been working to try and make happen, as I agree that concrete numbers like that are huge influencers. I’ve been reaching out to some people using the service to see if they would be interested in doing an actual case study with controls and metrics and have some interest from a few and am working to get things scheduled and coordinated. As an aside, you’d honestly be surprised at how many people pay for a service but don’t engage when you reach out to them. It’s odd, but that’s been my experience.

I read “Influence” and thought it was great. I saw that “Pre-suasion” came out and have been meaning to pick up a copy. I think I’ll treat myself to that tonight based on your recommendation :slight_smile:

Again, I really appreciate all this feedback. It’s been super helpful to get some different perspectives and to see that a lot of things I was thinking in the back of my head are things that others are thinking as well. Keep it coming!

Let me know how you like Pre-suasion. It’s got more anecdotes, so a longer read, but they’re really well written. BTW, I had no idea that Influence was like 25 years old!

It does not surprise me at all that someone would buy a service and then not want help with it. I think some people (many?) are afraid of talking to people. Perhaps they think it’ll be a sales pitch.

When trying to engage those people, approach like dating (abotu which, having been married 20 years I know only from observation)
Don’t ask for a date first. Just chat with them. Then ask for a very short (5 minute) conversation. Build trust. If you truly aren’t selling then they’ll realize that quickly an relax.

  1. Your service doesn’t need to be pixel perfect. 1% error rate (i.e., click within 2% of where it really was) is fine.

  2. How about a demo showing 4 radically different screens and what sort of heat map they’d get. (This is tricky b/c I bet a lot of people don’t realize the different screen sizes are a problem. If you want to kick a hornet’s nest post a comparison of yours to Crazy Egg.

  3. FYI, there is a way around the google “on page clicking” view : you give each clickable item a unique ID. (There is something I think called Extra Info or something where the click will include the ID of the element (or if none, it’s parent).

Yeah, that dating analogy is spot on. I agree a lot of people are wary of “strangers”, especially since almost all of the strangers they come across in business settings are the bad kinds of sales people that don’t really care about helping, but are really only interested in enriching themselves. I’ve been taking some different tacts to show that I’m genuinely interested in helping their website (and hence business) get better results and have been having some success with those approaches. I think you’re definitely right that getting that trust is key. It’s just amazing how conditioned everyone is by default and what you have to do to overcome that.

I like your suggestion about showing the different screens. I’m going to be putting something together with that shortly here.

Kicking that hornet’s nest is definitely appealing, but I want to make sure I can back up what I’m saying with a can of Raid before I pull that trigger. If I were to do that, I’m pretty sure that could turn out to be a really entertaining week. Better stock up on some popcorn first!

I nice feature I saw on one of your competitors, can’t recall which, was that they would use certain user behavior to recognize issues. Making it easier to identify recordings where users have usability or other issues.

Some of the behavior they used was:

  • Multiple clicks on or around the same spot (indicating navigational issues)
  • Lot of scrolling up and down (unclear content? )

There where more, but I think you get the idea. This really helps people sort through the videos faster.

1 Like

Hi Vlad, just wanted to let you know I looked into this and it looks like the page you were using behaves like a SPA (single page application). From what I can tell, you got an “unresponsive script” error because the event loop queue filled up. Support for SPAs (or really really heavy JavaScript pages) is on a “try it and see if it works” basis right now as I work through building test cases for heavy JavaScript/SPA scenarios.

That being said, I just double and triple tested Tamboo against my test pages in Firefox using performance tools and there’s nothing out of the ordinary, and definitely not any unresponsive script warnings.

I’ll keep you updated as I get closer to finalizing SPA support.