James Governor's Monkchips

Salesforce acquires Heroku: Dork Move, guys.

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

I came into the office this morning to see a tweet that stopped me in my tracks. Salesforce.com is acquiring Heroku for $212m. In cash! Beyond the obvious bubbliciousness of the deal was a cold hard truth about the once and future kingmakers – developers.

You see I didn’t first hear about Heroku from a marketing exec, or a PR person, or an analyst relations professional, I heard it from Tom Insam, a super-productive developer I was co-working with a little while before Dopplr was acquired by Nokia and airlifted to Berlin. This was very much a Silicon Roundabout story. Tom was all excited about Heroku because he could write code in Ruby on Rails and just deploy it to a server in the sky without needing to consider any implementation or operational details or capacity planning, or anything much. Completely situational.  Heroku was basically Rails on a billing engine.

Which brings us to Salesforce.com – which is basically a billing engine across a bunch of servers with some salesforce automation on the side. Obviously Salesforce is a lot more than that, but unlike many other players they always clearly understood that job one in utility computing was billing. Heroku – the same. Salesforce avoids IT to sell to the business, while Heroku avoids IT to sell to developers. The two firms definitely have something in common. While Salesforce has done an oustanding job selling to line of business people, its direct outreach to developers through its Force.com PaaS platform and “Java-like” APEX language has been disappointing so far. Big Difference then- APEX is “Java-like”. Heroku is Ruby.

Another thing Heroku potentially brings is some volume. Salesforce needs to track Amazon economics as a cloud provider, and Heroku runs on Amazon Web Services…. for now.

Heroku is currently spreading its wings, or should I say expanding the portfolio by looking to address the staggering opportunity around serverside Javascript in the shape of node.js. Why should you care about node.js? Because every developer in SF is currently going mental for it – its the emerging language for network services and services. Now interestingly enough Joyent recently acquired node.js outright, and its lead developer Ryan Dahl. Salesforce.com probably just added $150m to Joyent’s market value… But for now Heroku is all about the Rails.

What about Heroku – I can’t see the obvious evidence for a $212m evaluation in terms of customer success– although Best Buy is a nice name to pick up – but Salesforce is certainly playing an interesting game. Consider that Rails developers don’t much care what the database is, really (beyond MySQL and SQLite). But purchasers want assurance, and salesforce.com just introduced database.com as a backend it claims was designed for multitenancy – with salesforce’s 25bn transactions per quarter as a proof point for scale. So could database.com be the backend for Heroku? Why not? Lets developers develop, and business people pay for Quality of Service. Seems reasonable. Of course most apps will be situational and non-relational.

I don’t want to get carried away, and I haven’t even spoken to the parties involved yet, but price notwithstanding the fit makes sense strategically. Salesforce.com has all these services just waiting for developers to target them. Moving to RESTful interfaces was an important step, but the Heroku move radically ratchets up its profile with developers.

Talking of developers, my current office mates Matt and James tell me the new hotness in Heroku is the addon model for providers. This integration platfrom makes its stupidly easy for developers to sell their own services through Heroku. Say you want high volume email for your app – just use Sendgrid. No config- it *just works*. Or say you want application performance monitoring from New Relic. Again – one click and you’re there. Developers can call operational functionality without talking to ops. This is profound – and very much a marketplace. Its not exactly devops as a platform. Remember XML for config? Well now you can forget it…

Salesforce isn’t alone in unleashing its inner dork lately. Lots of other firms currently are- linkedin and Yahoo are good examples. This deal is really not about the consumerisation, so much as the developerisation of IT. Dennis Howlett got some nice quotage from Parker Harris at salesforce:

“We’ve got this Force.com platform and that’s amazing but what we really need is to make a connection to the vast number of developers who don’t know who we are or what we do.”

When IT executives talk about “making a bet” they usually just mean responding to market pressure, but in this case salesforce really is looking to make a market. No enterprise software company has yet nailed Platform as a Service.  Salesforce just invited the new kingmakers to the party in order to try and change that.

What is Heroku’s pitch?

Fast, frictionless, and maintenance free. Deploy and scale with confidence.

Sounds just like salesforce, right? We’re moving from no software to no hardware…

disclosure: Salesforce is a client


  1. If Heroku “is” something, it’s *Ruby*, not Rails. As you (hopefully) know, Rails and Ruby are not synonymous any more than Django and Python are synonymous, or Symfony and PHP, or Node.js and JavaScript. The frequent and persistent conflation of Rails and Ruby by bloggers and the tech press is aggravating to those of us who love Ruby, but aren’t as enamored with Rails.

    The distinction is very important. Heroku isn’t just useful to Rails developers, it’s useful to anyone writing server-based applications in Ruby, regardless of what framework they may be using. And while Heroku, to date, has focused on Ruby, they have plans and architecture in place to support other languages and platforms in the future (as you mentioned).

    I think it’s important to be clear about both of these points, especially in a discussion about Heroku’s value to developers.

    1. thanks for taking the time, Ryan. funnily enough i explained the distinction to @dahowlett earlier. points well made.

  2. […] little after I wrote up the news yesterday that salesforce.com was acquiring Heroku I came across a quote from Marc Benioff, tweeted by @timanderson “Ruby is the language of the […]

  3. […] James Governor says that Salesforce.com and Heroku share some similarities. He calls Salesforce.com a billing […]

  4. The future is (could and /or should be) event driven PaaS as opposed to waiting (or blocking) on some event. nodejs is that model.

    database.com is db via a network event.

    nodejs and database.com appear to be quite a fit. Obviously, Javascript (server-side) and Ruby are going to “cloud” the issue with more than one choice for apps.

    All the same, you seem to have a split personality when you tout IBM software and indirectly their way of licensing proprietary software licenses under so called enterprise license agreements that penalize advancements in hardware AND on the flip side you analyse salesforce, which is a “the best software is no-software” model, quite well.

    It almost seems that you need to make a choice on your alignment between your two paying clients.

    After all, the real good for society is buy-side analysis (which you open source) versus sell-side (which is driven by the TARP$ investment banks merger and acquisition advisory services).

    1. hey Derek thanks for taking the time. Honestly – i don’t expect large enterprise(y) on premise software to go away any time soon. we have coverered node.js and continue to do so, but lets just say the technology is going to mature before we see mainstream IT departments buying it. software ELAs are the waqy of the world – customers choose them. Should I ingore the reality of the market in terms of my analysis? I don’t think I do need to make a choice, in terms of my coverage, and certainly not based on who pays me. As it happens salesforce is a client but I’d have written the same note regardless. Heroku was the inspiration. We try and focus on the make side rather than the buy or sell side. We’re about practitioners whereever they live. Don’t forget open source – we have pushed that, often at the expense of our “paying clients”. But once again thanks for your comments- I think you get to the heart of how we live now. There are a lot of deployment choices out there, but existing development skills are methods aren’t going to be jettisoned, nor should they be. when you talk about penalise do you mean multicore?

  5. […] making a small strike against Amazon, the, er, overall kind of cloud. James Governor of RedMonk had an insightful take on why this $212 million deal was important: Which brings us to Salesforce.com – which is […]

  6. I agree that Heroku is a potential volume investment. I think they got em at a good price. Salesforce customers can now deploy and scale with MORE confidence 🙂

Leave a Reply

Your email address will not be published. Required fields are marked *