Why Tamr?

Over the past 2+ years, I’ve been fortunate to have the opportunity to contribute to the Cambridge/Boston start-up ecosystem as a mentor and investor in more than 25 local start-ups (list of investments here).


During that time, my partner and friend, Mike Stonebraker, was hard at work at MIT  on a number of research projects – one of which was in the area of large scale data integration and curation – the results of the research were published here: Data-Tamer.  Over the past 12-18 months, with close collaboration of some fantastic commercial partners, Mike, myself and a small group of world class engineers have been quietly working on a commercial project that was inspired by the Data-Tamer research.  We’ve been in residence @ Koa Labs and I believe the software we’ve been building will have a significant impact how how large enterprises use all their data to “compete on analytics.”  Our project has attracted some of the best talent in data systems in Cambridge – and for that matter from around the country.  One of our co-founders,  Ihab Ilyas, is an amazing talented computer science professor from the University of Waterloo.  Along with Ihab, Mike and a team of great people, we have launched the commercial project as Tamr today at the DataBeat conference in San Francisco.  (insert reference here to cliche Silicon Valley sitcom/companies that use two consonants at the end of their name – I guess this happens in Cambridge as well)


So it is logical for my friends to ask:


“Why in the world would you take what looks like a full time job now…after working so hard over the past 10 years to avoid a full time job?”


In short, I’m focusing on Tamr instead of just doing Koa Labs + Fund because I believe that we’re at a turning point in enterprise data management and analytics and there is a need an opportunity to build a significant independent company.  Mark Schreiber describes the motivating problem very well in this blog post.


Most large enterprises are generating a significantly wider variety of data every day and this variety challenge (more than volume or velocity – if your counting Big Data “V”s) is becoming the biggest bottleneck for enterprise analytics. The solutions and expertise that Tamr will provide to large enterprises over the next 5-10 years can create tremendous value for our enterprise customers as they digest a never ending number of diverse data sources in their quest to become more data-driven and begin to compete primarily on the quality and sophistication of their analytics.


High quality data from many sources is very much the fuel of analytic competitive advantage for the connected analytic enterprise. Tamr is like a high throughput refinery converting diverse raw data sources into high octane information fuel for analysts and data scientists to optimize every decision in the enterprise and help enterprises maximize shareholder value and achieve the mission of their organization. In short, the hype about big data and data science is great, but data science depends on high quality data fuel; this is exactly what Tamr is designed to deliver.  There are many great tools for empowering data scientists with statistics and visualization, but unless the quality and connectedness of the raw data is improved, even the most sophisticated front end tools will merely be window dressing on a spoiled carcass of bad and disconnected data.


There is always an ulterior motive – so what is it this time?  As Christopher Ahlberg has been telling me for more then years, the best way to non-incrementally improve Cambridge/Boston’s start-up infrastructure is to create more strong, independent companies that remain independent in the long term.  I’m very hopeful that Tamr can earn the right to become an anchor company in Cambridge that will help inspire the next generation of experienced entrepreneurs in New England.


I believe in full contact entrepreneurship, so Tamr has my primary attention for the foreseeable future.  As much as it pains me, I’ve had to put new investments on the back burner and divert my attention from spending time mentoring other entrepreneurs and start-ups to focusing on my own.   I have to admit it feels good to have my hands on the wheel again as I’m an operator – not a professional investor or advisor.  I will remain on a few select Boards but otherwise my focus is going to be exclusively on Tamr as we go through this next phase in the company’s development.   I hope you’ll wish us luck as we dig into the incredibly noisy data space (yea I know – crazy – right – imagine how a customer feels) and try to shake things up… again.


As anyone who has worked with me over the past five years knows – if you are a mission driven entrepreneur in Cambridge and need some help, just show up for breakfast at Tory Row and I’m likely to lend a hand wherever needed.


Check out some more on Tamr:
Aside | This entry was posted in Analytics, Big Data, Companies, Enterprise Software, Entrepreneurship, Founders, Information Technology, Life Sciences, Start-Ups, Uncategorized, Venture Funding and tagged , , , , , , , , , , , . Bookmark the permalink.

2 Responses to Why Tamr?

  1. Best wishes on this venture. My 2 cents worth of advice is to never forget that data analysis is merely “problem solving,” and that it is an art that is continually undergoing modification. It’s success is very much dependent on your ability to both process within & outside of your established logic and parameters. Utilize the vast advantages of technology memory for processing, and the advantages of the human brain for insight & intuitiveness in problem solving, as you design your new systems.

  2. Pingback: Founder Collective |

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s