I think Cakewalk's demise was simply down to money and management decisions. Sonar is a great DAW, I don't think there was any fault in Sonar itself to cause Cakewalk's demise.

In any product business, sooner or later you have to manage the balance between new sales and renewals. Once you get to the point where you've saturated the market (or your market share obviously isn't increasing), then setting yourself up to comfortably survive on renewals should be the goal. That way new sales are a bonus, but you don't have to rely on them.

Of course keeping your existing customer base, and ensuring renewal costs are at the right level is vital here.

Here are a few areas where I think is where things went wrong for Cakewalk:

1. Sonar X1: The workflow changed significantly here, and although it was arguably better, it took me a while to get used to the new interface and workflow. It was also pretty buggy. Not everyone would have stuck with Sonar at this point, i.e. if you're going to have to learn a new workflow anyhow, then why not try out other DAW's?

2. The "rent to buy" / "subscription" thing: I think the confusion and initial backlash over this put a lot of people off and hurt Cakewalk.

3. Lifetime upgrades: This was a novel idea, but it could only be successful if combined with decent new sales and plenty of paid-for additional content. This obviously didn't happen. You need sustainable income from somewhere: either it's from upgrades, or it's from new sales or content.

So the alternative would have been to stick with the old model, however looking at the figures around the time Roland took over Cakewalk, this obviously wasn't working either.

I think all of the above was done in good faith to try to turn things around, but it just didn't work out.

As a software development manager myself who has just semi-retired a product, there's a few things that maybe Cakewalk could have done:

1. Recognise that Sonar is a pretty complete DAW, and focus on bug fixes/stability rather than new features for the core product... at least for a a year or two in any case.

2. Focus on ease of maintenance by investing in automation, code refactoring and removing any deprecated code that is causing maintenance headaches. In my day job we managed to get a 130+ workforce working on one product down to around 6 people (with most of the other staff now working on new products). It took about a year to get it to that state, but now at least it's financially viable in maintenance mode - i.e. the support revenue more than covers the remaining staff costs.

3. For new development, maybe focus on added value than on the core product. These could be modularised and sold accordingly. Things like:
 - Integration with notation software, wave editing software, composition tools etc
 - Hardware specific modules (like say having the ProChannel leverage the internal DSP's of RME / Tascam interfaces) - SoftTube Console 1 is a good example, but it was way too expensive.
 - Better project migration/import/export from other DAWS. IMHO this would be a HUGE selling point for pro studios (especially ProTools import/export).

4. For future core development, take a step back from trying to innovate. Let the other DAWs innovate first, try it out, and improve on it. It seems to me that's what Cakewalks's competition have been doing for years, so try beating them at their own game.

The most important thing here however is knowing your customer base, balancing cost/return on investment, and getting your priorities right. There's no point in introducing features that won't be widely used or appreciated.

Just my 2p's worth...

M.

SandlinJohn

sharke
<snip>
 
Add to this the fact that no well known producers or artists are open about using it, and the fact that a negligible fraction of a percentage of the production videos you see online feature Sonar, and it's not hard to see why it was having so much trouble attracting new users. 
 
<snip>

Well, there is this: http://www.cakewalk.com/Artist
 
Of course you could argue these folks are not well known.

Is this the type of endorsee that might have helped?
 
https://www.steinberg.net/en/products/accessories/alan_parsons_dvd_tutorials.html 

Excellent post (#31) Mark (msmcleod)

SandlinJohn
Well, there is this: http://www.cakewalk.com/Artist
 

Hah, TC Spitfire was a bandmate with my son when he was just TC. I used to record him in the 90s.

SandlinJohn

sharke
<snip>
 
Add to this the fact that no well known producers or artists are open about using it, and the fact that a negligible fraction of a percentage of the production videos you see online feature Sonar, and it's not hard to see why it was having so much trouble attracting new users. 
 
<snip>

Well, there is this: http://www.cakewalk.com/Artist
 
Of course you could argue these folks are not well known.

And there are several in this list that are not only known, but legendary. Steven Slate regularly uses Cubase to demo his plugins.
 
https://www.steinberg.net/en/artists/steinbergartists.html
 
Cubase is proof that you don't need to be a loop-based daw to succeed. Does it do loop construction well? Yes. But, it also does a lot of other genres, and the features are extended compared to Sonar.
 
According to the Reddit thread, there were more managers than developers. Never a good sign unless it's something small like cranking out apps.

Hah Hah. That Zimmer photo is only about 20 years old.

I got the impression Cakewalk was trying to squash everything into Sonar.
You can only sell one Sonar, so to speak. Leave Sonar to its existing user base.
 
A new product that borrows the best ideas from current DAWs
would have been a real new product to sell.
It could have leveraged some existing code from Sonar.
Project 7?
 
But Cakewalk also needed reputation, visibility and market awareness.
 
I notice on the Ableton forums people complain so little has improved
but users are being asked to pay the next expensive upgrade again.
In other words, reputation and reality can diverge.
 
Marketing is important. And at Cakewalk, this was lacking.
 

Marketing.  There was none.

jpetersen
Leave Sonar to its existing user base.

 
That's kind of why it died. 
 

sharke
They really needed to spend a hell of a lot more on fixing existing bugs and quirks, many of which have plagued the program across multiple versions and given it a bad reputation. Its development team was no doubt stretched too far and things just weren't adding up. 

Yup. I lost count how many bugs I logged in the end. Only 3 or so got fixed and the rest of issues stayed open without anybody looking at them. Instead we got drum replacer, vocal align, and shiny. It DID get better, but bakers were not transparent enough. You would only hear from the bakers for a few days after release...and they would just pick their favourite issue and leave the rest hanging. Then you would have to listen to the Andertonian propaganda.
 
Anyway, yes who cares. Nice to see some maturity here about this. There was a time when if something like this was even mentioned a brawl would break out, every single time.