Big Data Driven Real Time Valuation – The next wave in Performance Management Systems?

A recent study suggests that the value gap between the “stock price based market cap” and the “internal cash flow based market cap” of Fortune 500 companies, are constantly on the rise, especially in the last 50 years. The implication is that some companies are either grossly undervalued (up to 25% points in case of old economy style companies) or overvalued (up to 50-100% points, in case of dotcom/entrepreneurial style companies) - and rightfully so, this value gap, seems to be emerging as one of the top 5 issues, keeping the CEO, senior leaders and investors, on the alert.

Let’s face it - the true stock value of a company, in its essence is the combination of its “future cash flow value” and its linkage to “future expectation building abilities” discounted by today’s cost of capital. While there are quite a few valuation models (e.g. McKinsey's Zen of corporate Finance formula, DCF, NPV, IRR) available, to accurately quantify the cash flow value, there are not many proven models available, to accurately establish the causal chain relationship between the cash flow value and the “expectation building abilities” - and so, stock markets, often follow the tread mill effect of trading on mismatched expectations.

Part of the reason for that behavior is that, the so called “expectation building abilities” are often hidden within the “intangible” competitive advantage (CA) enabling perspectives of the company (i.e. learning/growth, customer, value chain and financial perspectives in BSC terminology) and so, they are not well understood by the street, within its right context. “Expectation building abilities”, in this context include, but not limited to are - the earning guidance, product portfolio pipeline for the next 5 years, share repurchase program schedules/dividend payout, management changes and other big ticket capital expenditures - provided by company executives on an ongoing basis, to better manage street’s expectations, and to boost their overall stock prices.

Trading Value vs. Creating Value

The implication is that street, either over reacts to those “expectation building abilities” in the form of higher stock prices, or in some cases, under react (or some might suggest punish) with lower prices – purely based on the way, it interprets the causal relationship between cash flow value and the “expectation building abilities”. As it turns out, this causality interpretation gap seems to be the single most important factor, that drives many company executives and investors, to get into this game of trading value, as opposed to creating them fresh, says Roger Martin.

What do we mean? From investor’s standpoint – while quantifying cash flow is important in making the near term investment decisions, interpreting the causal relationship accurately, is key for them to bet on a stock for the long haul. Similarly, from company’s leadership standpoint – while quantifying cash flow value is critical for making effective near term operating decisions (i.e. achieving Q-To-Q results), framing the “expectation building abilities” accurately, in the form of earning guidance statements, is very important for them to manage the expectations of their investor community, and to stay focused on their long term strategic choices.

How about Valuation in M&A deal scenarios?

As it turns out, this causality gap between “expectation building abilities” and cash flow value seems to be the biggest hurdle faced by leaders within M&A negotiation scenarios as well – as suggested by a recent study. For example in a M&A scenario, if we had to dissect the value components of a product life cycle of a target firm -

• Invisible value of discovering the unmet “Jobs to be done” value + Invisible idea/vision seed value +invisible incubator value + commercialization value(when there is no alternative product available) + commoditization value (when alternate products are available with heavy competition)
  

If we look at this equation, interestingly enough, they map 1:1 to the five value stations of the PTV framework picture above. Yet another interesting insight here is that- it reiterates our earlier point that stock prices are often decided based on the commercialization and commoditization vale components (i.e. financial value station), partly because, street does not have visibility to the unrealized invisible values of the early product life cycles, the primary driver behind those expectation building abilities.

Transparency without compromising the Integrity of Insider information
The follow-up question is how can we provide that end-to-end visibility, without compromising the integrity of the CA enabling insider information? While there is never a silver bullet answer for this type of questions, one of the best answers in our opinion, is to mine those “expectation building abilities, proactively from the tons of data that are often hidden within and outside the four walls of the companies, and integrate them within the performance management systems. Yes, you guessed it correct - that the emerging buzz word for that approach is “Big data”.

As it turns out, even company insides, often do not have visibility to all of their own company data, leave alone the external social media data, and so, the better way to frame the question is – how can this emerging big data concept, help us to establish the causal relationship between cash flow and “expectation building abilities” and provide that end-to-end visibility, without compromising the integrity of the CA providing insider information?

Big data – the panacea for establishing the causality between cash flow and “expectation building abilities”?
Let us face it - the key benefit of any analytics platform (leave alone the Big-data driven analytics platform), is to help make effective decisions with timely and accurate insights. With that definition, if we had to dissect the service components of a Big data analytics platform, it is all about delivering value in the form of “insights” using a set of “analytics services” (as the delivery vehicle), with a faster service delivery time (i.e. latency) than the traditional analytics platform. In other words, the three key service components of big data are:

• Matter or cost in dollars that are needed to create and deliver those insights


• Acceptable Time or latency involved in delivering those insights


• Space or size of the data of the data needed- which is often why the word big is put in front of data.

Big data classified using CLS index
Now that we have defined Big data with the building blocks of matter, time and space – it makes sense to augment our firm’s Experience Pool Portfolio (EPP) framework with some of the value-add features of Big data, to accurately establish the causal relationship between the cash flow value and the “expectation building abilities” that are being hidden within the large amounts of Big data. Speaking of intangibles being buried within big data, the next question is – what is preventing us from quantifying those intangible values within the big data – is it a cost (matter), latency (time) or storage issue (size)?

The answer in our opinion is not using the right set of data sources for the right type of analysis, and so, as a first step, we suggest to classify “Big data” with a new index called CLS index (Cost, Latency and Size), as shown in the chart below.

Depending upon which quadrant the data falls, we see five buckets of big data categories evolving.

• Transaction data producing the value for financial station (structured data with a low CLS mix index)


• P&S Purchase context data producing the value within the value chain station (structured data with a medium CLS mix index)


• P&S Purchase influencer data producing value within Learning/Growth station (structured data similar to IRI/Nielsen data with above average CLS mix index)


• Customer needs & want based data or” jobs to be done” data, producing value within Customer station. (combination of structured and unstructured data with high CLS mix index similar to social media/survey data in the form of opinions, likes etc)


• Purpose or Motivational driver’s data producing value within the central purpose station. (some combination of above four categories, helping us to answer the key purpose drivers question).
  

Big Data comes to life within PTV radial framework
Interestingly enough – this five types of big data map 1:1 to the five value stations of PTV - suggesting the need for four additional virtual exchanges on top of the wall street exchange ( i.e. primarily financial perspective driven)– and link them all together with causality - as outlined in the picture above and explained in detail in our firm’s Purpose driven strategic planning framework (http://www.managementexchange.com/story/strategic-planning-purpose-driven-way-using-nature%E2%80%99s-seedal-chain-principle)

• Job to be done or customer equity capital exchange, based on the formula -> Consumer/customer Value=Jobs-to-be done/Price



• Human capital equity exchange, based on human capital learning/growth, per one of my recent hack at MIX site (http://www.managementexchange.com/hack/reforming-performance-management-systems-%E2%80%93-virtual-purpose-equity-vizpity%C2%A9-exchange-way).
  


• Value chain equity exchange,based on the formula ->ROIC= Margin x Velocity



• Financial equity exchange, which is today’s’ street version based on the formula ->Share holder value= Profit x (1-g/ROIC)/ (WACC-g) - as outlined in our firm’s valuation driven PTV radial framework below.



• Purpose equity capital exchange ,based on our hypothesized formula -> Purpose value= Profit x (1-g/ROIC)/ (WACP*-g) where WACP=Weighted Average Cost of Purpose Capital
  

Providing this type of valuation visibility in these five dimensions, not only will help us to accurately establish the causal relationships, but also, help us to make the right call in M/A deal situations, as we had explained in one of our CPM articles (http://theacademyofbusinessstrategy-businessanalytics.com/2010/05/15/03/)

Conclusion with Implications

While the solution we have proposed is a 18 month solution, let me conclude with some immediate pragmatic next steps, that can help companies to lay the foundation for implementing this Big data driven performance management systems. First and foremost, companies must identify their causal/correlation chain relationships between “cash flow value" and the corresponding “expectation building abilities” impacting their overall stock prices. Some of the practical steps include, but not limited to are -

1. Identify the heavy hitters (top 50-100 investors) who have the “needle moving power” to influence value (and thus stock prices) within this causal/correlation chain relationship using a CPM framework like our PTV framework, as shown in the picture above.



2. Proactively manage the expectations of family owned and hedge/pension fund investors and document what makes them to tick.



3. Establish the causal chain of how they have reacted to the earning guidance and macroeconomic news/factors in the past, and then create a behavioral pattern map for their behaviors.



4. Depending upon those patterns, manage their expectations by releasing right type of insider information to media (and the analyst/investor community) in the right time, with a news staging mindset.



5. When company’s stock price changes, ask why the market moved and zero-in who bought, who sold, and why - with an empathetic mindset i.e. look at those investors with a lens of “alter-ego managers” or “indirect corporate owners”. In other words, learn to empathize with investors with various “what if scenario” options, within the context of the five perspectives of PTV framework.



6. Overhaul investor relations department and make them to actively manage the five big data types, by making them as joint data stewards, especially when it comes to regulatory data and annual reports.



7. Administer the big data registry with a causal map of shareholders and investor road shows, visits by analysts, and conferences; and major presentations to shareholders.



8. Integrate investor relationship management, part and parcel of strategic planning. Together make them responsible for managing the key-account processes to identify movers and understand their behavior. This way one can test all major plans, hypothesis and announcements, for their effect on the price of the company’s shares with various “what if scenario” options, and then suggest modifications to those earning guidance content, to better align with the views of key shareholders, thus becoming the key advisory arm of the CEO and senior leaders/board of directors.



9. Make investor relation leaders (in partnership with Corporate Strategy and Finance) as the people who co-own this big data driven valuation system and make them to proactively deliver the bad news when necessary. They will also have to be experts with “thinking on the feet” type communication skills, capable of handling tough interviews with investors who at times might be pressing them for information that cannot be divulged under SEC regulations or for maintaining CA integrity.




10. In closing, this type of big data driven integrated approach (i.e. structured and unstructured data driven approach) to strategic planning, Innovation, performance management, investor relations/valuation/stock price management and risk management, clearly require top notch talent, including the time and attention of senior management - and failing short on any of those commitments, would definitely leave the CEO, senior leaders and board of directors, in the constant game of never ending stock price guessing. It is a no brainer that no CEO and/or senior leaders/ board of directors, would ever want to be in that type of a guessing game, and so, the million dollar next step question is “What is in your docket ?” – and let that be our last word!

Sensitivity Analysis Framework for Improving Profitability and Valuation

Some of the ideas we have been promoting in the last few weeks – appear to have caused some buzz within the blogosphere- more specifically, our sustainable valuation formula, definitely has caught the attention of few readers (http://strategywithapurpose.blogspot.com/2010/11/purpose-profit-balanced-sustainable.html). One of our fellow readers made a casual observation - “our simplified valuation formula [i.e. V= NOPAT x (1- g/ROIC)/ (WCCP-g)] does not seem to take in to account the impact of all key value drivers (e.g. pricing, margins, volume, COGS etc) –and, hence it trivializes the whole valuation exercise”.

While we agree that it is a simple formula, it is an accurate one from financial algorithm standpoint. The reason we took McKinsey’s “Zen of Corporate Finance” as the base in our article last week was -to build a simple yet compelling case for discounting the free cash flow with a purpose value driver (WACP) very much like how WACC is used to discount the free cash flow. At the same time, we definitely empathize with our readers –who, otherwise perform the detailed valuation exercise with multiple spreadsheet pages of DCF cash flow projections and value driver assumptions. We still recommend our readers to go through the similar DCF process and validate the result with our summarized formula. In other words, our formula is more of a validation formula than a working formula.

The question however is – does that mean our sustainable valuation formula do not take in to account the impact of those granular value drivers? Before answering that question, let us first dissect ROIC within the valuation formula

ROIC = NOPAT/Sales x Sales/Capital => Operating Margin x Operating Velocity

As we further inspect the Operating margin and its sub components– some of the key value drivers that contribute towards the operating margin are – Price, COGS and Volume - which are further dependent upon promotion scenarios (including their depth and frequency), competitor responses, elasticity and boundary conditions to name a few. We sure can expand our formula mathematically to explain how these value drivers are indirectly accounted for within our valuation formula. However, it does not communicate the causal chain relationship effectively, and so we thought that it is a worthwhile exercise to come up with a sensitivity analysis framework to show the impact of these value drivers on margin using a real world case study – clearly showing the impact of price changes on volume (& hence on margin and the overall NOPAT/enterprise value).

Our goal for this case study is to develop a flexible framework model from the methodology we had used for one of our client CPGR-Co Inc. (our fictitious CPG/Retail Corporation pronounced as ZipGyarCo) who had asked us to help them to increase the profit by 10% (& hence enterprise value) by increasing (or decreasing) price by a certain percentage, yet without substantially impacting the volume.

The Working Framework Model

As part of that exercise, we decided to develop a working framework model (and a mathematical formula) of linking all the input and output variables that contribute towards increasing the operating margin, profit and valuation as outlined in the schematic at the top of the page. With all things being equal (consumer purchase criteria, brand perceptions, trade promotion rates and efficiency etc), we quickly realized that there still exists few cost differences (e.g. supply chain/distribution costs) across the market regions and hence we decided to develop a region specific pricing model (as opposed to one mammoth corporate model) to better simulate the real world business conditions. Within that approach, we then divided the CPGR-Co’s markets in to 5 RMA’s (East, West, Central, South and North) selling all of their top 5 Product models or package sizes (P1, P2, P3, P4, P5) – resulting in 25 model cells (i.e. one model each for each cell). Please note that the more granular we divide the markets the more accurate our models will be.

The objective of our case study was to develop an analytics based pricing model with a regression based mathematical formula producing the following projected results.

• Recommendation to price up or down in each cell.
• Projected volume, profit, share and valuation outcomes from price changes.
• Optimization recommendations for competitive response scenarios.

To achieve these objectives - we went through a systematic problem solving process with the following steps.

1. Calculate VCOGS matrix – at the product model level for each RMA.
2. Arrive at the optimum adjusted trade price (from retail price) to better negotiate with retailers/vendors.
3. Load the Product/RMA matrix with historical volume and price data from IRI/Nielson.
4. Develop the analytics based CORE Pricing model with all the inputs, outputs and their interdependent relationship.
5. Develop the equivalent log-log form mathematical equation simulating the core model- more specifically, develop a formula to arrive at the unit volume – which is a function of
   • Current price (e.g., P1)
   • Previous period’s price (e.g., P1_PriorPeriod)
   • Price of our other pack and/or model sizes (e.g., P2, P3)
   • Price of our competitors (e.g., CPCp1, CPcp2,C Pcp3, CPcp4)
   • Brand Equity impact
   • Other ancillary variables (e.g., Holiday, Trend etc)
   

The corresponding equation format is

New Volume = e0 + b1*ln(PP1) + e2*ln(PP1_Prior_Period) + e3*ln(PP2) + b4*ln(PP3) + …
e5*ln(CPCP1) + e6*ln(CPCP2) + e7*ln(CPCP3) + e8*ln(CPCP4)…
e9*Brand Equity + 110*Holiday …

Where

• e1 - own price elasticity (negative) - “If we raise P1 price by 1%, P1 volume will change by e1%”.
• e2 - lagged price impact (positive) - “If we raise P1 price by 1% this period, next period’s P1 volume will change by e2%.
• e3 and e4 - own model size cannibalistic -elasticities (positive) - If we raise P2 price by 1%, P1 volume will change by e3%.
• e5-e8 - competitive elasticities (positive) - “If competitor raises CPcp1 price by 1%, P1 volume will change by e5%.
• e9 - Brand Equity impact (could be negative depending upon the depth and frequency of the promotion and where the firm is on the innovation maturity curve) - Every additional period we move forward in time, we expect P1 volume to change by e9 * 100%.
• e10 Holiday impact (positive) - For holiday periods, we expect P1 volume to be e10*100% higher than for non-holiday periods”.

6. Calculate the price elasticities for three scenarios (SELF, SELF CANNIBALISTIC and COMPETITIVE) based on IRI/Nielson data.

7. Adjust the results based on competitor responses.

8. Adjust the results based on boundary condition analysis.

9. Use economic model to further optimize the results across various regions/accounts.

10. Finalize the projected volume, profit, share and valuation changes from price changes.

Value Driver Strategy Execution Considerations

As we can recognize from the steps above, we can easily extend this working model and mathematical formula for other value drivers like COGS containment, inventory cycle reduction, idle capacity containment (& few more) as well for arriving at a similar set of recommendation/results for improving profitability and valuation. However, one of the big challenges in implementing those recommendations within large organizations like “CPGR-Co” is building the consensus across various stake holders – from the standpoint of - first identifying the right set of appropriate value drivers, developing the right framework, arriving at the the action plan and finally implementing them.

For example, although there is a substantial opportunity to achieve profit/valuation goals by making the pricing changes as identified by our analysis - executing those price changes in a timely manner, is not an easy thing to do- given the fact pricing is embedded in a wide range of broader business decisions like strategy development (e.g., capacity utilization, new product pricing), price optimization (e.g., average price targets) ,trade strategy and sales execution (e.g., EDLP vs. HiLow) scenarios. In addition, the optimal pricing decisions require the input and expertise of a wide range of domain “experts” from both business units and front line sales strategy teams.

To circumvent these execution challenges, we made a recommendation to our client “CPGR-Co” to form few cross functional Value Driver Councils (VDC’s) to coordinate and proactively shape these value driver decisions (e.g. pricing decisions) across accounts, channels and brands. More specifically, within the context of our value driver council for pricing, the VDC will be chartered with the following

• Define pricing strategies/moves across the portfolio.
• Identify other value drivers that compliment pricing.
• Communicate approved pricing strategies to the field.
• Ensure region/account-level trade policies fit with strategy.

In addition, create a forum or platform for regular review of pricing trends, and other value driver needs by analyzing

• Market trends, competitor behavior
• Performance of new products
• Performance of competitor products

Finally, also act as the steering committee or governance layer for long-term pricing improvement

• Identifying “next level” pricing challenges
• Pricing pilots/tests—spreading best practices
• Upgrading analytics foundation (tools, processes, techniques, policies and organization models)

“Winning by Analytics” Considerations

As we can clearly recognize from the framework model steps and VDC’s governance layer requirements, Analytics is critical for successfully executing the tasks in every step of the way. We cannot stress the importance of analytics enough - that we encourage organizations like CPGR-CO to develop a solid analytics foundation with a right set of tools, techniques and templates to reach the winning level of "Level 5 analytics maturity" as promoted by Davenport & Harris and as explained in some of our articles published by Academy of Business Strategy (http://theacademyofbusinessstrategy-businessanalytics.com/). Within the context of that winning spirit, we specifically encourage our client CPGR-Co to develop the following set of tools.

• RMA/Product Matrix level optimization tool, techniques and templates to develop tailored pricing decisions across pack and model sizes and RMAs.
• IRI /Nielson price elasticity tool, technique and templates to estimate profit and volume impact of these decisions across competitive scenarios.
• Average price to trade calendar tool, techniques and templates to estimate impact of trade calendar scenarios on average price/unit.
• Category management tool to develop compelling cases to support field execution of pricing decisions.
• Forecasting and sales management tool to provide input to financial management system and to track progress against pricing targets in the form of “what if analysis based” drill down dashboards.

Bottom Line:

At the end of the day – developing winning strategies is definitely a great thing to do – but, institutionalizing them is all the more important - as strategy in its "execution form" is what brings the tangible outcome to stakeholders. Let us face it - fully empowered price strategy decisions (or for that matter, any other value driver strategy decisions) requires extensive consensus building in most organizations- and this is where- our Value Driver Councils (VDC) come in to the picture. So, it is a call to action for our client “CPGR-Co” and other similar clients-not only to develop winning strategies to improve profitability (and valuation), but also, institutionalize them using VDC’s.