Let the data scientist who is without sin cast the first algorithm

The chances are that we have all been guilty of failings in our big data past but, how do we ensure we are not beyond hope of redemption in our analytical future?

1) Greed

‘The pursuit of material possession’

Analytics success is a process of mutual not individual value-creation.

Your big data endeavours will rarely succeed if immediate material impact is a personal reason for commencing the initiative. The side effects of greed include an unwillingness to spend what is needed to get the job done, as well as the sin of hoarding accrued insight in the mistaken belief that success centres on individual achievement.

Sometimes the intelligence derived in the early stages of an analytics project is priceless in the long-term without obvious bottom-line impact in the short-term.

Sometimes the intelligence derived in the early stages of an analytics project is priceless in the long-term without obvious bottom-line impact in the short-term.

2) Envy

‘The desire of anything that belongs to your neighbour’

Competitor or peer analytical success is not necessarily portable.

The belief that we can port an analytical outcome that was successfully delivered elsewhere is a fallacy. Big data solutions are unique to each business, its culture and challenges. Sure, we can learn something from the approach and innovation of others but only as a guideline rather than a prescriptive route to success.

Loose the ‘petabyte envy’ and drive your solutions against specific known challenges within your organisation. Often it is those who have the most money and resources that fail hard through focus creep.

3) Wrath

‘The pursuit of self-destructiveness and feud’

Real data power comes from ‘open innovation’ rather than our chequered history of closed 'command and control' hierarchies.

Not every big data cycle will succeed and, when they do it’s a team play not an individual competition. Resist the temptation to pick faults with the solutions of others just because they may not align to your thinking and approach. Data is less powerful when we are unprepared to freely innovate and adapt. Think ‘reuse’ first then control and governance second during periods of extreme analytical innovation.

Remember, the network always wins in this age of uncertainty. Have an open mind and let a variety of data-sets change the team mind-set with trust.

4) Sloth

‘The failure to do the things that one should do’

Data ‘apathy’ by business and IT is terminal for the organisation

Doing the right things for the business often means doing the ‘wrong things’ for our data and technology landscape and vice versa. It seems easier for data professionals to hide behind a veil of technology governance than to solve the real problems of an imperfect world. Furthermore, business owners should not see data as a technology discipline. A lack of proactive governance and ownership of data across our core processes causes insurmountable problems down the line. Culture in this respect kills innovation before it starts.

Analytical success is directly proportional to our collective flexibility to do the ‘right things at the right time’ across front-office and back-office operations with pragmatism rather than idealism at heart.

Analytical success is directly proportional to our collective flexibility to do the ‘right things at the right time’

5) Gluttony

‘The overindulgence and over-consumption of anything to the point of waste’

Hoarding information assets and project funding is a recipe for disaster

Too often an unwillingness to share one's data expertise is mistakenly believed to be the path to job security; have data professionals heard of the sharing economy? Equally, saying ‘Yes’ to every analytical request without benefits and achievability assessment to artificially ‘pool funding’, ensures that delivery teams end up running at emergency levels for numerous misguided project outcomes in parallel. The path to enlightenment lies with ‘our data, right focus’ rather than 'my data, unlimited scope'.

You can analyse ‘all’ of the data ‘some’ of the time and, ‘some’ of the data ‘all’ of the time but, you cannot analyse ‘all’ of the data ‘all’ of the time. Not yet .........

6) Lust

‘Uncontrolled desire at times of temptation’

Big data success requires emotional focus not relentless spend.

The most common expression of analytical lust is the endless pursuit of bigger and better technology simply for the purpose of ‘new technology’. Often, what we have is more than sufficient for the task at hand if sensible scope focus is applied. Career aspirations are more likely to be met when we deliver focused ‘bite-sized’ business benefits at minimal outlay rather than significant upfront investment in ‘interesting data toys’ as a pre-cursor to actually starting the analytics benefits journey.

Look to adapt what you have first. Pilot using express, trial or open technology if you must but, deliver a ‘prototype’ outcome to depict the ‘art of the possible’. Investment will follow once the value is proven.

7) Pride

‘Failing to acknowledge the accomplishment of others’

Embrace rather than replace the analytical perspective of others

Pride manifests itself where business and IT professionals believe they know ‘everything they need to know about organisational data’ – the reality is we each only truly understand a ‘slender’ perspective. Success lies in listening to the insight others can provide, admitting cultural mistakes in how we harness data plus, ensuring we are prepared for success or failure at each project iteration.

So what is my path to analytical redemption?

If we think of big data projects as an iterative cycle of development:

>Accelerate Delivery from the outset.

Gain the necessary momentum for your big data initiative by removing unnecessary barriers to progression.

>Avoid technology lust, outcome greed and performance wrath.

Focus on an achievable pilot scope, sweat your existing technology assets and share the outcome benefits across multiple beneficiaries.

>Openly empower the wider business.

Get a wide variety of subject matter expertise involved early and do not be afraid to develop ideas using less than ideal technical approaches in the early stages.

>Avoid peer envy and secular pride.

Design the best outcome for your organisational within the confines of your existing budgets and technology assets. Do not allow individual subject matter expertise of the data estate diminish the potential innovation of others.

>Improve and expand continuously

Adopt an approach of 'ever increasing circles' to not only improve the quality of your intelligence but also, to widen the variety of its insight.

>Avoid agility sloth and iteration gluttony.

Ensure that each expansion of scope is both rapid and manageable and, that it continues to widen both mutual benefits and the speed of actionable insight.

Read. Repent. Repeat and then go forth and not only multiply but divide, extrapolate and predict with a clear conscience

By Simon Gratton:

Simon Gratton (@Simon_Gratton) is an experienced CDO, CTO, business strategist and data ‘artist’. He has a history of driving business change and innovation using information as THE strategic asset, having honed his experience most recently at Zurich, Capgemini, Royal Mail Group and TELUS. With a passion for data at the heart of the next generation customer experience, Simon continues to build transformative data cultures using an innovative blend of technology, process, recruitment and training approaches which range from government apprenticeships and university collaboration to sharing economy hackathons. Simon is a frequent keynote speaker and a respected industry ‘blogger’. Know more about him here.