Make Analytics Useful, Meaningful and Actionable

Last week, I identified reasons for the organizational malady of failing to fully leverage analytics to make higher quality decisions in less time.  As promised, this week, I want to share a remedy.

For the analyst, I recommend the following:

  1. Put yourself in the shoes of the decision-maker.  Try to step back from the details of your analysis for a moment and ask yourself the questions he or she will ask.
  2. Engage your decision-maker in the process.  Gather their perspective as an input.  Don’t make any assumptions.  Ask lots of questions.  They probably know things that you don’t know about the question you are trying to answer.  Draw them out.  Schedule updates with the decision-maker, but keep them brief and focused on essentials.  Ask for their insight and guidance.  It may prove more valuable than you think.
  3. Take time to know, explore and communicate the “Why?” of your analysis – Why is the analysis important?  Why are the results the way they are?  To what factors are the results most sensitive and why?  Why are the results not 100% conclusive?  What are the risks and why do they exist?  What are the options? 
  4. Make sure you schedule time to explain your approach and the “Why?”  Your decision-maker needs to know beforehand that this is what you are planning to do.  You will need to put the “Why”? in the context of the goals and concerns of your decision-maker.
  5. Consider the possible incentives for your decision-maker to ignore your recommendations and give him or her reasons to act on your recommendations that are also consistent with their own interest.
  6. “A picture is worth a thousand words.”  Make the analysis visual, even interactive, if possible.
  7. Consider delivering the results in Excel (leveraging Visual Basic, for example), not just in a Power Point presentation or a Word document.  In the hands of a skilled programmer and analyst, amazing analysis and pictures can be developed and displayed through Visual Basic and Excel.  Every executive already has a license for Excel and this puts him or her face-to-face with the data (hopefully in graphical form as well as tabular).  You may be required to create a Power Point presentation, but keep it minimal and try to complement it with Excel or another tool that actually contains the data and the results of your analysis. 

Frustration with your decision-making audience will not help them, you, or the organization.  Addressing them where they are by intelligently and carefully managing the “soft” side of analytics will often determine whether you make a difference or contribute to a pile of wasted analytical effort. 

Thanks again for stopping by.  I hope that these suggestions will improve the usefulness of your analysis.  As a final thought for the weekend, consider these words from Booker T. Washington, “There is no power on earth that can neutralize the influence of a high, pure, useful and simple life.” 

Have a wonderful weekend!

Advertisements

Why the Soft Side of Analytics Is So Hard to Manage

I’m borrowing both inspiration and content from two good friends and long-time supply chain professionals, Scott Sykes and Mike Okey.  They deserve the credit for the seminal thoughts.  Any blame for muddling the ideas or poorly articulating them is all mine.

If you are an analyst, operations researcher or quantitative consultant, you probably enjoy the “hard” side of analytics.  What we often struggle with as analysts is what you might call the “soft” side of analytics which is always more challenging than the “hard” stuff.  Here are a few of the reasons why.

Many times, the problem is not insufficient data, defective data, inadequate data models, or even incompetent analysis.  Often, the reason that better decisions are not made in less time is that many companies of all sizes have some, if not many, managers and leaders who struggle to make decisions with facts and evidence . . . even when it is spoon-fed to them.  One reason is that regardless of functional or organizational orientation, some executives tend not to be analytically competent or even interested in analysis.  As a result, they tend to mistrust any and all data and analyses, regardless of source.

In other situations, organizations still discount robust analysis because the resulting implications require decisions that conflict or contrast with “tribal knowledge”, institutional customs, their previous decisions, or ideas that they or their management have stated for the record.  Something to keep in mind is that at least some of the analysis may need to support the current thinking and direction of the audience that is analytically supportable if you want the audience to listen to the part of your analysis that challenges current thinking and direction.

Understanding the context or the “Why?” of analysis is fundamental to benefiting from it.  However, there are times when the results of an analysis can be conflicting or ambiguous.  When the results of analysis don’t lead to a clear, unarguable conclusion, then managers or executives without the patience to ask and understand “Why?” may assume that the data is bad or, more commonly, that the analyst is incompetent.

Perhaps the most difficult challenge an organization must overcome in order to raise the level of its analytical capability, is the natural hubris of senior managers who believe that their organizational rank defines their level of unaided analytical insight.  Hopefully, as we grow older, we also grow wiser.  The wiser we are, the slower we are to conclude and the quicker we are to learn.  The same ought to be true for us as we progress up the ranks of our organization, but sometimes it isn’t.

So, if these are the reasons for the organizational malady of failing to fully leverage analytics to make higher quality decisions in less time, what is the remedy?

The remedy for this is the subject of next week’s post, so please “stay tuned”!

Thanks for having a read.  Whether you are an executive decision-maker, a manager, or an analyst, I hope these ideas have made you stop and think about how you can help your organization make higher quality decisions in less time.

A final thought comes from T.S. Eliot, “The only wisdom we can hope to acquire is the wisdom of humility—humility is endless.”

Have a wonderful weekend!

Guest Post: Big Data Is Getting Bigger! Are Retail Companies Ready?

This month’s McKinsey Quarterly carries an article, “Are You Ready for the Era of Big Data?”.  In contrasting two competitors, it states the following:

“The [One] competitor had made massive investments in its ability to collect, integrate, and analyze data from each store and every sales unit and had used this ability to run myriad real-world experiments.  At the same time, it had linked this information to suppliers’ databases, making it possible to adjust prices in real time, to reorder hot-selling items automatically, and to shift items from store to store easily.  By constantly testing, bundling, synthesizing, and making information instantly available across the organization—from the store floor to the CFO’s office—the rival company had become a different, far nimbler type of business.”

This week’s guest post explores what retailers need to do in order to take advantage of “big data”.  The author is Özgür Yazlalı, one of my former colleagues and a gentleman who has years of experience helping retailers significantly improve decisions that directly impact financial results through data-driven analysis.  I greatly appreciate Özgür’s contribution and look forward to more in the future.

_______________

As an internal consultant in a Fortune 500 retail company, one of the interesting transitions I have witnessed was the significant change in all dimensions of available data.  Day by day, not only more granular data but also more categories of data have become increasingly accessible to planning functions.

The retail store remains the closest link to the end customer of supply chains – the consumer.  Leveraging this connection requires collecting tremendous amounts of very detailed transaction data.  Consider the following examples:

– Customer:  “For this very random reason, I’ll rightfully return this product that I don’t remember from which store I bought”.
– Retailer:  “Ooops, we cannot see how much we charged you for this product, so we’ll credit the ticket price back to you.”

The emergence of social electronic media has multiplied the complexity and volume of interaction between consumers and retail stores.  Smart phones, social networks, image processing, and cloud computing have become reality in our everyday lives, further illustrating the relentless and ubiquitous nature of Moore’s Law.

The “big data” phenomenon is a great opportunity for the manager who knows how to utilize it.  The possibilities of localized assortment planning with reliable store-product category forecasts, localized pricing, and demand planning with social networks and web-based trends are now very real.  The limitation lies neither in the analytical capability nor the existence of reliable data, but rather in the legacy planning systems.

This phenomenon can also be a curse if all you know about data is limited to spreadsheet programs.  My humble observation is that there is now an ever-expanding gap between the rate of increase in the data and the spreadsheet capabilities of individual users.  Getting the big picture could now be a difficult task.  Without a better way to address this gap, you risk being limited to analyzing specific events and/or drawing conclusions based on a limited sample.

Given the increasing gap, the traditional way of building IT solutions based on business requirements documents and restricted interaction is no longer viable.  Big data requires cross-functional and data-capable analytical teams that operate as intermediary functions between business and IT organizations.  This is not a team simply put together from ex-business and ex-IT folks, but data scientists, optimization experts, experienced data and business analytics consultants that could unleash the capabilities of SQL and spreadsheets together.  Such teams not only facilitate the discussions between the two organizations but also enhance a tool design process that is more interactive with rapid innovation and prototyping.  This is critical because no one really knows a priori what assumptions and models will consistently work.

Most companies are organized as silos of functions, as are their data sets (though IT might store these datasets in the same server).  Thus, in addition to enhanced IT-business interaction, these analytical teams could work with different business functions that do not routinely communicate, leveraging data as common ground.  For example, the inventory management teams that use sales and inventory data may not know about, or may choose to ignore, trends in store traffic that consumer insights teams usually track.  Yet, an integrated analysis of these two data sets could reveal that a decreasing sales trend in a well-inventoried store is being driven by poor assortment, even if there is sufficient traffic.

Big data presents significant challenges and opportunities to businesses.  Cloud-based, functionally local solutions may help individual business functions maintain their own small data warehouses, but an holistic approach demands greater IT involvement.  Unhappily, IT organizations are not often seen as centers of innovation.  A team of skilled, experienced analysts with access to both the computational power owned by IT and the big data in which other business functions find themselves awash provides an effective means for overcoming the challenges and delivering on the opportunities of big data.  These analytical teams can be a part of the IT organization, but whatever organizational structure is employed, they must have powerful computing resources, access to all relevant data, and a voice with decision-makers. [Editorial Note:  A hyper-performance, secure, could-based platform would be the natural vehicle for data of all types from all sources where challenges and opportunities can be identified, diagnosed and the next best action determined and directed.]

In summary, big data is now more accessible, and companies must continuously explore new ways of using it to increase profitability and market share.  For retailers in particular, these opportunities are far greater than for any other industry because of both their proximity to consumers and the availability of structured data (social networks, CRM, POS, inventory, traffic, e-commerce).  Integrating all this valuable information into a predictive [Editor’s note:  and prescriptive] planning process is a difficult task requiring tighter linkage between the IT and business functions.  Analytical consulting teams with enhanced data-capabilities who could facilitate and guide this interaction are now more important than ever in achieving this goal of increased profitability and market share.

____________________________

Thanks again for dropping by.  If you liked this week’s guest post, please rate the piece and feel free to leave a comment.  Until next week, remember the words of the American poet, Harry Kemp who said, “The poor man is not he who is without a cent, but he who is without a dream.”

Have a great weekend!

%d bloggers like this: