A Wave of Acquisitions in Business Intelligence

Holy crap! Within the past week we’ve seen the acquisitions of the two biggest players in the modern BI landscape, Looker (announcement) and Tableau (announcement). And if you broaden your view to the entire analytics tech stack, it’s bigger: here are the major acquisitions I’ve tracked over the past year:

• Stitch. Acquired by Talend on 11/7/2018 for $60m.
• Alooma. Acquired by Google on 2/19/19 for an undisclosed amount.
• Periscope. Acquired by Sisense on 5/14/19 for an undisclosed amount.
• Looker. Acquired by Google on 6/6/19 for $2.6b.
• Tableau. Acquired by Salesforce on 6/10/19 for $15.7b.

Compare that to the list of widely-used products that remain independently owned / maintained:

• Snowflake
• Fivetran
• Mode
• Chartio
• Redash
• Metabase
• dbt (full disclosure: I’m one of the creators)
• Matillion

Of these, Snowflake is the 800-pound gorilla, certainly worth over $1B. The others vary significantly in terms of maturity but are all widely deployed.

So!? Why is this happening? What does it mean about the future of the stack? Here are my thoughts.

1. Enterprise Exerts “Gravity” on the Entire Stack

The push towards enterprise has been long in coming. Mode switched its pricing from $19 / user / month back in 2016 to a negotiated contract. Looker’s base pricing has continually edged upwards, as has Fivetran’s. Stitch’s enterprise business has taken off.

This isn’t necessarily a bad thing — it means that enterprise buyers are now interested in buying the same tech that us startups have been using for a while now. The danger to startups is that these large dollar-value contracts could crowd out smaller companies with smaller budgets as vendors change their pricing models.

This crowding-out is happening, but not universally. All of the major data warehouses still have self-service trials and pay-as-you-go models. Mode still has a free single-player mode. Stitch’s self-service options are very attractive for small businesses. And dbt is open source. So it is still very possible to build a high-quality modern data stack as a budget-constrained organization. But the overall gravity of the enterprise is pulling all vendors in that direction.

I experience this every day in my role at Fishtown Analytics. Open source users are a tremendous marketing channel for us and self-service subscribers are a critical part of our flywheel—we care about both of these user groups deeply. It seems increasingly likely, though, that large enterprises will make up the better part of our revenue in the future.

As large organizations migrate to the modern data stack, it’s inevitable that the vendors in the space would cater to them. And it shouldn’t be surprising that Big Tech wants to acquire those same vendors once they attract this enterprise attention.

2. BI as the Thin Edge of the Wedge

I have no insider knowledge about this, but here’s my theory on the Looker acquisition. Google recently hired Thomas Kurian to run GCP, and smarter people than me believe that this is all about improving GCP’s position in the enterprise.

As someone who’s now done many hundreds of sales calls selling the integrated analytics stack, I’ve seen just how critical the BI layer is in convincing buyers of the value of the entire stack. My personal opinion is that the underlying tech—the ingestion, transformation, and warehouse layers—actually matter more than the BI layer, but buyers rarely see it this way. Buyers in this space buy the front-end. It’s how they, and how most of an organization’s users, interact with the underlying technology.

As such, Google likely wants Looker because it’ll act as the thin edge of the wedge for larger GCP deals. Looker has real traction in the market but is currently only monetizing the BI layer. If instead that traction can get monetized through the entire stack (which Google now owns tools at all layers of), ACV grows massively.

Of course, Looker will continue to play nice with the non-GCP parts of the stack. But you can bet that Azure and AWS won’t be solution-selling with Looker as much any more, and Microsoft and Amazon don’t own an asset that is even remotely competitive with Looker. PowerBI and QuickSight are not on the same playing field with Looker when it comes to an enterprise-wide deployment.

Looker will absolutely help GCP sell more GCP.

3. Open Source: The Next Wave of BI?

Much of the story that’s played out over the past 7 years reads very similarly to how the previous round of that story played out in the 90’s. Upstart technology vendors get some traction, sell to increasingly large companies, and get gobbled up by incumbent platforms. The 2012–2019 version of the story isn’t quite over yet, but with the speed of acquisitions in the past year (and past week!) it’s certainly headed that way.

It’s interesting to compare this story with the story of companies like MongoDB, Elastic, and Redis. These companies make similarly critical enterprise technology, but are all built on top of open source software. These companies have demonstrated significantly more staying power as independent entities than the companies within the modern analytics stack. I believe that their open source foundations help create that longevity: open source communities are tremendous engines of growth and yet can be challenging to monetize for acquirers.

The reason I think this is interesting is that there may be seeds germinating for a coming wave of open-source BI. Redash and Metabase have huge communities of engaged users. Apache Superset’s creator, Max Beauchemin, is rumored to have raised VC to create a commercial offering for it. And dbt continues to grow its install base aggressively. Each of these products is early, and many lack important features that exist in closed-source competitors. But they also have tremendous traction, momentum, and community love.

It’s too early to make a strong statement here, but this is an important trend to take note of. If it is true that enterprise software companies built on open source are more resilient to acquisition, it’s conceivable that an open source-led round of innovation might promote more long-term innovation than another wave of proprietary software companies who would inevitably become BigTech’s next meal.

4. Competition Will Shift to Providing New Experiences

The BI tools from the modern analytics stack all had fundamentally the same go-to-market plan: build a fairly traditional BI tool on top of a more powerful data warehouse. The innovation in this space kicked off initially because of Amazon Redshift — Redshift and its successors Bigquery and Snowflake are so powerful that they created an opportunity for new fast-moving software vendors to rebuild the BI landscape on top of them.

But the core experiences weren’t fundamentally new. Write some code, configure some charts, get a dashboard, share via email. The core experience of using these tools is at the same time dramatically better than, but not structurally different than, the prior generation. They’re better because they’re more user-friendly, they’re web-based, and they plug into better underlying warehouses.

As BigTech comes to dominate the “modern” BI companies, I think that the next chapter in the story will have to look fundamentally different. A part of this will likely be open source (see above), but I also believe that startups will come up with fundamentally new user experiences to deliver on top of these fast-evolving data warehouses.

Here are some places I’m looking forward to tracking in this area:

• Automatic schema discovery empowering the business user. The modern stack primarily sells to the analyst, could metadata + ML help empower the business user to self serve start-to-finish?
• Rich interfaces for exploring event data. Existing BI tools just don’t shine on top of event data yet it represents an increasing proportion of the data acquired by companies.
• Delivering amazing KPI dashboards. Looker, Mode, and Periscope are fantastic for lots of use cases, but still pretty “meh” when it comes to creating a high-information-density executive dashboard. What if a tool focused specifically on this?

These are just what come to the top of my mind — I’m sure there are others.

I’d love to hear your ideas in the comments — who do you think are the winners and losers here? Who’s going to get acquired next? What does it all mean for the future of the industry, users and vendors alike?

Edit: I was made aware in the comments that there was another big acquisition in February: Qlik acquired Attunity for $560m.