Tech firms know a lot about you…

but that’s nothing compared to data brokers, who collect from everywhere and sell to anyone.

A Friday series exploring Startups and the people who make them go. Read all If the Shoe Fits posts here.

The following post is reprinted in full with Wally’s blessings. I e-met Wally when we both blogged for b5 Media — I think. It’s so long ago I’m not sure, but over the years I’ve read and appreciated Wally for both his insights and independence from accepted leadership-speak. I highly recommend adding his blog to your reading list.

Quantitative Data and Self-Deception

If you need someone to blame this on, it might as well be Rene Descartes. The 23-year-old Descartes was serving in the army when was visited in a dream by the Spirit of Truth who told him that “Conquest of nature is to come through number and measure.”

Numbers were power. That effect was amplified during the Industrial Revolution. That’s when the engineers took charge, measuring and calculating. Soon, Frederick Taylor and the efficiency experts showed up with their stopwatches and clipboards. Now we’re in the Digital Age, where computers spit out numbers by the mountain load.

Today, companies trumpet the claim that they’re “data-driven.” The Economist proclaims that “data is the new oil.” If there is a golden calf to worship today, it’s probably digitized.

We love numbers so much that we don’t think about where they come from or how we’re using them. We can summon them from our vast databases, manipulate them, and turn them into equations that give us “answers. It makes us feel like we’re in control. We’re not, really. We’re only in control of the data.

The map is not the territory and data is not reality

Data is not reality. At best, data can only represent reality. Reality is complex and messy and we can use data to simplify parts of it so we can understand it better. To do that we must leave out part of reality, assign numbers to things that aren’t inherently quantifiable, and approximate relationships with equations.

If, after all that, we treat data like reality we commit what Alfred North Whitehead called “the fallacy of misplaced concreteness.” We get lost in the wonder of our calculations and think we’re describing the elephant, when we’ve only got hold of one leg.

It’s a good idea to apply George Box’s observation about models to our data. All are flawed, but some are useful.

Quantitative data is not objective

No matter what you or your boss thinks, quantitative data is no more objective than qualitative data. Someone, somewhere, sometime decided what would be counted and tracked and what would not. Someone, somewhere, sometime decided how and how often data would be gathered and how it would be presented.

That’s obvious when you talk about qualitative data. We usually get qualitative data in the form of a story. This happened when we observed it this way. With quantitative data, the questions, assumptions, and decisions that lie behind the data are usually behind the curtain and invisible to the people who receive and use the data.

Dig into the history of things to find out why you use certain measures and not others, how the raw data is gathered and manipulated, and why it is presented in the way that it is.

Quantitative data is not enough

Quantitative data is important, it’s just not enough for a successful business or a satisfying life. The most important things in life and business can’t be counted or calculated. Relationships drive much that happens in business. More than half a century ago, Mason Haire demonstrated that emotions influence buying decisions of all kinds. Knowledge workers trade in conversations and tacit knowledge.

There’s one more thing about quantitative data. It’s easy for us to manipulate and “understand” quickly, so we’re likely to pay attention to what we can count and ignore what we can’t. That’s part of the reason why the long term is often sacrificed to the short term and why numerical accounting data gets more attention than “soft” human stuff. As one friend of mine said years ago, “When the pressure’s on to make the numbers, people almost always take a hit.”

Bottom Line

Quantitative data is important. You can’t run a successful operation today without paying attention to it. Remember that quantitative data is always a flawed representation of reality. Look behind the curtain to discover the whys and hows behind the data. Remember that human choices drive quantitative data as much as qualitative data. And, please, remember that the most important things in life and business cannot be force-fit into a dataset.

Copyright © 2017 Wally Bock, All rights reserved.

Image credit: HikingArtist

How would you feel if someone constantly followed you and then shared that info with friends?

Would it bother you more if the info was sold for cash?

Would you report your stalkers? Or at least find a way to stop them?

Essentially, that’s what Google does.

It follows you on your jaunts around your cyber-world and both shares and sells that info.

Remembered the last time you surfed around looking for a particular product and then found ads for the same thing on every page you looked at for months afterwards?

What many of us consider commercial stalking Google and others call “improving the user experience.”

For decades, our Congress, in its infinite wisdom, has pooh-poohed the idea of any kind of privacy policy, such as Europe has, saying it would hamper growth.

My solution is using the DuckDuckGo search engine that doesn’t track you, or for total anonymity I use ixquick.

But what can you do if you’re addicted to Google and have been using it for years?

You can say thanks to Business Insider and use the step-by-step, illustrated instructions for deleting your history preventing continuing surveillance that they recently provided.

The funny thing is that what most people want is choice, i.e., the ability to easily opt out when a search is extremely sensitive — by their definition, not a third party’s.

And, at the end, since it’s all about money, perhaps if enough people opt out Google will change its approach and give you a simple way to decide who is privy to what in your own little corner of cyberspace.

Or, an even more heretical idea, pay you for it use.

Image credit: E Photos and Business Insider

A Friday series exploring Startups and the people who make them go. Read all If the Shoe Fits posts here

As a founder, do you have an ethical or moral responsibility to consider the ramifications of your product on society — globally, not just locally?

In Numbers, a TV show that ran from 2005 to 2010, the central character, Charlie Eppes, was a young, prodigy mathematician. One storyline forced him to question his long held belief on his responsibility in innovation.

I always believed it was my duty to develop numerical tools and someone else’s to use them wisely.

Is that what you think?

In your drive for sales would you sell to any who could pay or would you scrutinize them to assure ethical usage?

Some companies do just that.

Data analysis powerhouse Palantir has been ultra-careful since it was founded.

Palantir can afford not to sell to just anybody — you have to believe in its values, too (…)  Palantir once turned down a partnership with a tobacco company “for fear the company would harness the data to pinpoint vulnerable communities to sell cigarettes to,” CEO Alex Karp told Fortune.

Jad Saliba, Magnet Forensics’ founder/CTO and ex-cop is emphatic on the subject.

 “The two areas I care most about are combating terrorism and child exploitation,” he says, adding that he hopes to keep his company on the side of the angels. “We spend a lot of time validating who we sell to … We sell to people who are going to use it ethically.”

Big Data in all its forms has enormous potential for good — and even larger potential for abuse.

And AI even more so.

From man’s earliest days, every new discovery has been a two-edged sword — fire can bestow life or death.

And while the final, future outcome of an innovation can’t be predicted, it should still be the responsibility of its creator, whether individual or company, to work to assure whatever it is is used responsibly.

Image credit: HikingArtist

This post is for all the fact-loving, data-crunching guys who keep claiming that tech is a merit-based ecosystem where anyone with a good idea who is willing to bust their tail 80 hours a week will succeed.

If you are one of them you probably aren’t aware that March is Women’s History Month; a time to celebrate women’s accomplishments, especially in tech, since they are why you have a company/job.

How excited would you be if it took 10 years for your most important metric to double?

That’s what you see for founding teams with at least one woman — from 7.7% in 2006 to 17.5% today.

Whoopee.

It’s much worse for all-female founding teams — their funding dropped from 22.8 in 2014 percent to 18.9 percent now.

That totally sucks.

And it’s far worse when you add color to the equation.

What’s it going to take for this to change?

More female angels and VCs — happening very slowly.

More angels and VCs of color — a distant dream.

But more importantly, and hopefully sooner, more successful, entitled white guys will digest the numbers and decide it’s just plain wrong.

 Are you/will you be one of them?

BTW
Happy St Paddy’s Day to all my Irish and Irish wannabe readers!

Image credit: TechCrunch/flickr and Free-extras

his year’s Vator Splash Health, which took place at Kaiser HQ in Oakland, was Startup focused and well worth the price of admission.

As we’ve come to expect, it featured a very impressive line-up of panelists and speakers dealing with extremely relevant topics ranging from opening remarks (kidding), tackling cancer with technology, to patient-centered healthcare, to telemedicine and patient engagement, to protecting yourself as the founder, to uncovering new data from API’s and platforms, to big picture items, such as the future of healthcare altogether.

There were supercharged Splash competition presentations featuring three of Health Tech finalists creating an opportunity for new businesses to effectively message their product.

Participating vendors were easily accessible; including Bloom Technologies, DocDelta, Lighthouse, from the American Diabetes Association, Lab Sensor Solutions, Carrum Health, and Crediyo.

Event partners included KPMG, SAP Startup Focus, Bread & Butter, Artis, Scrubbed, Stratpoint, and Healthiest.

On the agenda throughout the day were fireside chats with the likes of Helmy Eltoukhy, founder & CEO, Guardant Health and Vator founder/CEO Bambi Francisco.

Other splash talk topics showcased — when software eats bio, funding the science behind healthcare, Who’s financing the digital health ecosystem?, and Are You in Kaiser’s Line of Sight: A Buyer’s Perspective.

Big data was discussed at length, crystallizing the notion that it is the current ability, made accessible by modern technology, to put meaningful patterns together that are deployable that will affect outcomes and achieve objectives.   

An additional topic or two that I was pleasantly surprised by was the acknowledgement by Dave Schulte, Managing Director at McKesson Ventures, of the importance of the virtue of humility, in admitting “not knowing”. Kudos to Dave because this, of course, comes against the backdrop of Silicon Valley’s famed hubris. 

Leading to another interesting point in that, at a minimum, the possibility (if not certainty) exists, that there will be a falling away or clearing of many of the startups and downturn both in investment activity and new business creation.

A sober but fair assessment and reminder of the unavoidable cyclical nature of business that correctly tempers expectations. 

More than simply being a fun event, populated by interesting shakers and market makers, with good,  healthy food (a very pleasant change), it was a phenomenal networking forum and that, perhaps, is its most intrinsic value.

How freely do you discuss the details about how you think, what you like, what you believe and the challenges you face with strangers?

Sites, apps, data brokers and marketing analytics firms are gathering more and more details about people’s personal lives — from their social connections and health concerns to the ways they toggle between their devices. The intelligence is often used to help tailor online experiences or marketing pitches. Such data can also potentially be used to make inferences about people’s financial status, addictions, medical conditions, fitness, politics or religion in ways they may not want or like.

How willing would you be to sell that information to benefit a total stranger?

What if it would benefit a pet company, such as Apple, Facebook or Hulu?

You already give up your personal information in return for better access to their products and services, but you do so with the idea that you won’t be packaged and sold.

In fact, most sites tell you upfront that they won’t “share your personal data with third parties.”

But, as they say, the devil is in the details and buried deep in the privacy statements is a giant ‘but…’

Of the 99 sites with English-language terms of service or privacy policies, 85 said they might transfer users’ information if a merger, acquisition, bankruptcy, asset sale or other transaction occurred, The Times’s analysis found. The sites with these provisions include prominent consumer technology companies like Amazon, Apple, Facebook, Google and LinkedIn, in addition to Hulu.

It’s a safe bet that if these sites have that caveat, so do thousands of others — both large and small.

The expansion of the Internet of Things provides companies a far more intimate look at your life than ever dreamed possible.

It’s a trend that is likely to widen as companies introduce new Internet-enabled products, like connected cars and video cameras, which can collect and transmit a constant stream of data to the cloud.

Your best hope (if you care) is to assume that caveat emptor reigns.

Generally, caveat emptor is the contract law principle that controls the sale of real property after the date of closing, but may also apply to sales of other goods.

Your data is ‘other goods’.

Stuff happens; economies go up and down and businesses wax and wane.

Any company, no matter how large or seemingly stable can find itself in the position of having to sell or transfer its assets.

Your data is an asset. Period.

Flickr image credit: safwat sayed

The Data Alchemy Conference that I recently attended was well worth going to. In contrast with a lot of these types of conferences, it was an interesting view of how to use predictive and other technologies to improve business outcomes, i.e. not the more common type of technology or data scientist oriented conference here in Silicon Valley.

One of the factors that was attractive was the way that vendors used case studies and best practices to elucidate some of the advantages and complexities of big data and analytics. People from companies such as PayPal, IBM, HP, SAP, Silicon Valley Data Science and others were speakers. There were also lots of industry practitioners in the audience.

The emergence of predictive analytics as a core tool in planning and monitoring in organizations is a relatively recent phenomenon, being less than 10 years old. Now, companies like SAS have been around for a long time, but it is only when IBM acquired SPSS in 2009 and applied their significant marketing engine behind predictive analytics that this market started to take off. 

Of course, it had been used with regards to risk analytics in insurance, churn analysis in telecoms companies and credit worthiness analysis in FICO scores, etc.

Since then we’re seeing predictive analytics being incorporated in many different areas in enterprises based on the growing amount of data and the increasing need to make decisions based on data.

This comes partly from increasing complexity in the business world, greater binary behavior (1 major company in each market that is 10x larger than #2), speed of growth and decline of companies, and decreased cycle times.

One of the most interesting talks was by Jenny Dearborn, Chief Learning Officer at  SAP, who spoke of the way they’re using predictive analytics with regards to employee turnover and onboarding. By using big data analytics on structured and unstructured data, it is possible to understand employee sentiment, training needs and likelihood of staying at the company.

A major challenge to analytics is data quality, what in common vernacular is termed bad or dirty data. Theresa Kushner, VP Enterprise Information Management at VM Ware mentioned that 1/3 of her staff were focused on data quality and cleansing.

It seems as if data quality is an even more important issue than being able to apply advanced algorithms to the data, and that by just ensuring that data is clean we can make better decisions that reduces the need for advanced algorithms in many situations.

In short, it was interesting to see how analytics is being advanced within organizations and getting a practical view of what challenges are faced from a business perspective.

FinTech, the wedding between finance and technology, is a hotbed of startups and innovation, especially in London. Now it’s lighting the fires of the investment community in Silicon Valley, so I prevailed on Ajo Fod, who knows the FinTech world well, first as a quant and now as an entrepreneur, to attend the Trading Show and share his observations with you.

I had the pleasure of attending Trading Show West Coast 2015: West Coast’s leading quant, automated trading and big data event last week. This is one of the most legitimate trading shows I’ve seen and truly geared to professionals.

The first thing that caught my eye, was the surprisingly large majority wearing business attire; I was expecting some confusion. Google tried to hold down the fort of casual-at-work and a few people were dressed in jeans, with long-sleeved shirts for good measure.

But finance won over West Coast causal even in San Francisco. My decision to dress in a brown suit and a tie was just the right measure down from full business dress.

I was impressed by the balance between different groups of professionals. Quants / traders / investors / hardware / risk management and students were all well represented.

Different scales of enterprise from startups to micro hedge funds to medium sized funds, such as AXA Rosenberg, to industry titans like BNY Mellon Financial and Blackrock were there, too.

The mix of speakers, from hardware tech providing fast access to markets to macro thoughts from Lex Huberts, was good, especially considering the audience.

Systematic trading and HFT is no longer about the fastest execution. The marginal advantage from trading faster needs to be weighed carefully against the cost of the infrastructure, while the ability to forecast farther into the future is significant.

Apparently, the fastest access to markets is provided by Algo Logic. They sell machines that race the path from tick data reception to placing trades in 1.2micro seconds!

They achieve this by storing the logic in hardware in FPGA (field-programmable gate array). They include trading logic and risk checks on the chip to achieve this kind of reaction time.

The speed is used to grab favorably priced orders before anyone else can. The winners at any speed tend to be the ones with higher algorithmic sophistication. The direction of development in this field tends to be about adding computing power to the FPGA.

The discussion on Co-location vs Cloud Servers focused on the tradeoff between speed and algorithmic sophistication.

Pravil Gupta of Quadeye Trading and Bert Shen from SuperMicro are both suppliers of HFT technology. The difference is that one is about more sophisticated but still very fast trading while the other is at the higher speed end of the spectrum.

Speed is not everything in the HFT world. The incremental speed edge costs significantly. While there will always be fast traders that grab obviously mispriced orders over a short time horizon, others will play the game of taking the not so short-term bet.

The roundtables covered a list of varied topics. As expected the round table audiences in the Bay Area were largely focused on state-of-the-art in Big Data and deep learning.

These technologies could be the future, but I don’t see as much profitable application of these technologies as there is hype.

FinTech startups seem to be numerous in data services for the finance industry. iSentium: works on estimating the sentiment of tweets. Another works on interpreting SEC filings. Strategies are being fed information faster to produce more efficient markets.

The past was a speed race. The future is going to be about more information used in smarter ways.

For example, Alpha Sangha, my startup, combines information from a variety of data-sources using complex models/algorithms that maximize profitability while filtering out noise.

Acronyms come and go, so here are three relatively new ones stay aware of.

BRIC : Brazil Russia India China
MINT : Mexico Indonesia Nigeria Turkey
ESG: refers to the three main areas of concern that have developed as central factors in measuring the sustainability and ethical impact of an investment in a company or business

Ajo Fod is the founder of Alpha Sangha, which helps companies optimize complex forecasting models or algorithms based on large quantities of past data while avoiding the common pitfall of noise. They can further increase profitability by mining for model/algorithm variants that are better fits based on historical data.

Ajo previously worked as a quant at BGI/Blackrock and Mellon. He has masters degrees in both Computer Science (AI) and Operations Research (optimization). He earned a BTech degree from the prestigious IIT-Madras.

I always look forward to attending events produced by AlwaysOn. They do an exceptional job bringing together high-profile players appropriate for the conference subject, entrepreneurs, service providers and other interested parties.  The Silicon Valley Innovation Summit 2014 I attended last week was no different, but the devil was in the details.

Those both present and presenting were recognized tech movers and shakers—well worth listening to—the networking was excellent and I made some stealth contacts I’m not at liberty to discuss.

Subject matter centered on mobile any/everything, Big Data, SaaS, subscriber, consumer and investment globalization, which left me a bit disappointed even though big data is Quarrio’s ’thing’.

There was no mention of the tech I’m hearing/reading about daily, i.e., artificial intelligence, nanotechnology, synthetic biology, etc., and the combination of these technologies with mobile and big data.

We all know that this kind of focus and talk follows the money, so I am left with a question.

Are the ideas being funded yesterday’s instead of tomorrow’s?

KG Charles-Harris is CEO of Quarrio and a frequent contributor to MAPping Company Success.