All businesses struggle with understanding the progressing demand climate around a product line. It is critical for businesses to keep on top of where and how, their products are performing. This is so they can maintain an efficient pipeline and make appropriate product line decisions. Strategic product line decisions are vital in maintaining low cost, maximized profit and optimal production efficiency.
Brief: On Existing Methods for Determining Product Line Strength
When I searched for other articles on this topic, all I found were articles advising on how to value a product before its release. There was no real notion of how to value the market for a product on a progressive or continuum basis. That I know of, there are five primary methods for valuing the market for a product after release:
- Brand Potential Index (BPI) – This is probably the best of the current methods. Involves the strength of a brand given its geographic location and sales volume. It can be used in conjunction with spatial (locative) analysis and penetration pricing. Uses geo-specific market demand to optimally price and strategically place products. Used in competitive analysis.
- Aggregate Planning - involves the evaluation of Demand, Business Processes, and Inventories to yield disaggregated products like: Management Resource Plan (MRP) and Master Production Schedule. Used to compare the expected demand against logistics/production costs to find an optimal allocation strategy for a particular market.
- Sales Mix Variance – Uses CVP analysis to calculate an expected contribution margin from which yields expected sales. These expected sales are expressed as a percent but is used to analyze the performance of a product as dictated by its profit contribution to the whole (see Pareto Chart). Then each product is compared against each other to come up with the optimum mix.
- Cost-Volume-Profit (CVP) – Involves the comparison of different product lines through their respective contribution margin which, is a function of sales price, sales volume, and costs (fixed and variable). Then each is compared against a break even point.
- Marginal return calculations – All “marginal return calculation” means here is, the marginal analysis done on the target rate of return. This target rate of return can be anything the company perceives as a profit indicator.
While some of the methods above are cohesive in their approach, none of them present product value on a continuum. Meaning, all of them are only point in time analyses. This can give businesses an unclear picture in the true strength of a product line across periods. If the business simply mis-times their calculations, a business could believe that their product is consistently under-performing. Potentially leading the company to the conclusion that a product line is unprofitable as opposed to just following a down trend. Lastly, unlike the RSI, none of the better methods (i.e. the BPI or Aggregate Planning), present their results graphically. In my experience graphs and charts can be extremely useful in understanding trend and strength movements. The hypothetical relationship between the RSI and its root values is a complicated one. To provide some context, it may be helpful to think of the relationship between the RSI and the underlying asset as one of an index and its components. The RSI like an index changes with changes in their underlying component, as such the RSI can show either a moderated or extreme view of the component change. The difference between the RSI and an index, is that the RSI displays the variance of in its components as a function of averaging over time. While, the index displays a degree of change as a measure of the weight and change of its components, without explicit regard, for time or duration.
What is the RSI?
This is why I propose using the RSI in product line analysis. The RSI or Relative Strength Index is used in technical analysis of securities to measure the strength in buying and selling patterns. The RSI is known as a technical momentum oscillator, or a trend strength chart. The RSI operates in such a way that displays the strength of an underlying asset, typically a stock, over a given period set. It does this through a unique averaging technique which applies weight to previous periods while treating the current period nominally. In the end, the RSI in stock analysis, simply, examines the number of days that a stock increases or decreases. Finally yielding a number, between 0 and 100, that will be closer to 100 if there are more gaining than losing days and visa versa.
Below is the stock chart for Dell Inc. with the RSI on the chart below:
In the above chart, the RSI can be found below the main box-plot graph. In the RSI chart there are two straight lines indicating the upper and lower bound of the RSI. When the RSI crosses these lines it can indicate a bullish trend(upper bound cross) or bearish trend (lower bound cross). These two lines form a basis for what would be considered a relatively equivalent pattern of buying and selling the stock. In other words, when the RSI crosses the Upper Bound it is likely to be over-bought. When it crosses the Lower Bound the stock is likely over-sold. These bounds are generally accepted to be equal to 70 and 30, respectively. These bounds can be changed but typically aren’t, as they are considered the “standard” specification. But, I will talk about this when I discuss its use as a product line strength tool. The last important fixed line to note is the middle line. This middle line is set at 50 on the RSI chart and unlike the previous two cannot be changed. This is because, in the event that the RSI does not cross the 30 or the 70 line, a downward cross of the middle line could indicate a downward trend and vice versa. The middle line is important as it serves as a reference to the position of the other points and the 50 line, just as in the 70 and 30, is considered a signal line. Meaning, that to cross any of the lines should “signal” a reexamination of the chart for interpretation.
How the RSI is Calculated
The RSI is calculated as a function comparing average loss to average gain. If you are unsure of how to do this calculation, please refer to this web page. It does a good job of explaining it the formulaic approach to the RSI. I will only cover how to calculate it in the context of the RSI and any confusion on the basic process should refer to this site. At this point, all that it is really important to note at this point, is that the RSI is a comparable average measurement tool. Which, compares the average loss to average gain in each period to determine whether something is “over-bought” or “over-sold”. This allows the RSI to measure the relative strength of a value given the difference of previous values, over a set period of time. This allows the user to not only compare the strength in value, but also the trend, as I will show later.
The formulas of the RSI are as follows:
Now, when I discuss the RSI as a Product Line Strength Value tool I will use an adapted form of these formulas. Please, just consider that as it is usually presented and the purpose for its formation, was for its use in stock valuation. I adapted it by first, considering the functionality of the variables as they relate to each other and their function in the formula. As long as the relationship was consistent the formula could be adapted for other uses beyond the scope of its original intent.
As an example of this “adaptive rationale”, consider a scenario in which you have, a set of 20 values between 0 and 1. In this case, when comparing the different values, an average value of .5 represents the midpoint. The same is true of a second set of 20 values 10 years from now with values between 50 to 100, a value of 75 is the mid point. What is important to note hear is that the RSI does not calculate based on an entirely absolute average but rather an isolated one. This prevents the RSI from being to contingent on historical value. Instead the RSI compares the averages of values over a specified period of time. This causes the effect of large changes to be neutrally adjusted over time. In other words, to the RSI, the value sets of 0-1 and 50-100 are, implicitly, equivalent. So long as the change from one to the other, has occurred gradually. As such, significant divergences in the RSI are meant to specify dramatic change as different from normal deviation.
Below, I established a hypothetical scenario for a company called “Widget Co.”. In this case, Widget Co. is testing the RSI on its product line strength using an integrated profit indicator and a selected period of averaging.
The Case of Widget Co. – Assumptions
Goal: Widget Co. is pursuing a higher level of adaptability and strategic influence given the nature of the product market.
Product: Widgets, a valuable program that is sold by subscription with an optional 7 additional features.
Profit Indicator/Current Method: Value product line through a gross margin calculation each month product has been available.
Production Efficiency: Complete production efficiency – Units Sold = Units Produced
Date Range: Jan, 15th 2011 – October 15th 2012.
Sales Price per Unit Schedule:
- Original Widget (Vanilla Subscription) = 71-73 per month
- Original+ (Add 1 option) = +3 per month
- Original Super (Add 2 options) = +5 per month
- Original Super+ (Add 3 Options) = +6 per month
- Super Duper (Add 4 Options) = +7 per month
- Holy Widget! (Add 5 Options) = +8 per month
- The Slammer (Add 6 Options) = +9 per month
- God Mode (Add 7 Options) = +9 per month
Sales price per unit range = ($71-$82)
Median Sales Price per unit range = ($74-$80)
Cost of Goods Sold Per Unit (2 fixed costs and 2 variable costs):
- Server space which is constant at 30 dollars per month (Fixed Cost)
- IT maintenance costs 15 dollars per month (Fixed Contract Cost)
- Wage expenses 12 dollars per month (Variable Cost)
- Customer Support which can range anywhere from 4 to 16 dollars per month
Cost of Goods Sold per unit range = ($61-$73)
Median Cost of Goods Sold per unit range = ($68-$72)
Widget Co. RSI Charts and Notable Features
It is important to note that increasing the RSI period also can lead to a drop in volatility. Where this could be helpful is it provides a much better case for the long term due to the relative containment of severity. As a consequence, however, the signals are not likely to be as strong and harder to track. On a similar note, as mentioned earlier, it is possible to vary the upper and lower bounds of the RSI. What these bounds illustrate, their root purpose, is to establish standard deviation points from the center. Almost like a control. So, when the RSI crosses that line it is meant to indicate the crossing of the trend outside of normal operations to beyond the standard deviation. This is why the 50 line is vitally important. To serve as a reference for the measurement of deviation. One might adjust the lower and upper bounds if they, for instance had lower risk tolerance. In other words, they wanted an increase in the signaling behavior so as to have a more active approach to evaluating their product line. In the end, deviation can skew results to the companies preferences, in terms of risk or whatever. So long as adjustments to these bounds are set to have an equivalent distance from the center.
Below are a series of charts and explanations regarding the RSI for Widget Co.
Chart 1a:The RSI (3), Gross Margin and Linear Forecast Trendlines
- Divergences in the degree of change between gross margin and the RSI (3). Particularly on May-12 and Sep-13.
- Indicates is that for the month of May the decline in gross margin is much more significant that the simple decline in value would suggest.
- Relative period strength is considerably lower than the average strength over the past three periods.
- A similar case for the Sept-12 point in which, the increase in gross margin is not as strong as it would seem due to the significantly lower average values in the previous periods.
- The slight divergence in the linear forecast line that the strength of the product line is increasing but to the same degree as Gross Margin.
- The benefits of using the RSI in this case, are:
- A clearer picture of the trend in Gross Margin is visible for the next three periods through looking at the RSI as it tempers the implicit level of volatility in Gross Margin.
- Prevents from radical strategic decisions that would have been made using Gross Margin alone. Such as a discontinuation of the product offering during a period of Gross Margin decline. If Widget Co. were only to view the change in Gross Margin, it would have an incomplete picture of the trend in the chosen profit indicator.
Chart 1b: The RSI (6), Gross Profit and Linear Forecast Trendlines
- The main purpose of this graph is to show how an extension of the period (from 3 to 6) decreases the volatility in the RSI.
- Also note, as stated above, the RSI does not necessarily need to cross the 70 or 30 to indicate a trend but rather, only needs to cross the 50.
Chart 1c: RSI (3), Median Sales Price per Unit and Median Cost of Goods Sold per Unit
- If, instead of Gross Profit, the profit indicator of Widget Co. was one of its components such as Cost of Goods Sold or Sales Price.
- Visibly apparent that there are many cases in which, the RSI performs inversely to one or both.
- It is also notable that in some instances it appears that the RSI is, in fact, predictive of the change in Sales Price or Cost of Goods Sold.
Interpretation and Conclusion
As mentioned earlier, in charts for the RSI, there are two bounds the Upper Bound at 70 and the Lower Bound at 30. Crossing these bounds traditionally indicates that a stock is overbought or oversold, but since being “overbought” or “oversold” does not necessarily apply to Gross Profit as it does stocks. It must be determined what crossing these bounds indicates in the context of Widget Co. So, let us consider that, since Gross Margin is arithmetically defined as:
= (Sales Revenue-Cost of Goods Sold) / Sales Reveue
The two potential causes for the RSI crossing these bounds must be attributed to the inverse relationship of increasing/decreasing, between the Cost of Goods Sold and Sales Revenue.
In considering this situation, I will not speculate as the exact nature of any incline or decline in the RSI. As the optimal conclusion, would be formed through a careful comparison of the trend and growth in all three lines.
However, as a sample interpretation, I will examine the occurrence of an Upper or Lower Bound cross when there is a decline in Cost of Goods Sold with an increase or no change in Sales Revenue.
In this scenario, a crossing of the Upper or Lower Bound, without any other information is likely to indicate a cost effective strategy or an period of unusually low costs when the RSI is above the 70 and a period of excess costs or an inefficient cost per unit strategy. Although, it the must be noted that where it may appear a firm is advantaged by staying above the upper bound. Maintaining the same level of costs, even if it was the cause originals for a higher gross profit, will still result in the lowering of the RSI. This is because the RSI only considers a particular period as stronger than the previous if there was significant change in the components, in this case gross profit. In other words, if crossing the 70 was caused by an a decrease in cost to 1/2 the previous level. Then maintaining the new level cost, does not guarantee the same strength of the product. That is almost the beauty of the RSI, it does an incredible job of simulating an environment of unknown factors like demand, inflation, interest rates, competitive pricing, etc. As such, the RSI is arguably, a superior metric of determining product line strength. Due to its ability to not only, present the chosen metric on a continuum which, can be interpreted for patterns, trends and comparative values to other metrics. But, also to properly illustrate the factors in a virtual error term. It also has a very nice graphical display but that is not so much an argument as a preference.
The RSI does have weaknesses. For one it can only support one value variable. That is not to say you could not make multiple from different criteria, but the criteria, in order to be used in the RSI must culminate into a single variable.
Another weakness is the variance in interpretations. Since it is a completely new concept, as far as I know, to apply the RSI to business process strategic situations it is hard to say what exactly a firm should or should not do. For example, trying to determine whether crossing the upper bound is a good or a bad thing. Although, this could be seen as a double edged sword in that it treats both increases to revenue and decreases to costs the same. Sort of like Moneyball where, walks are as good as hits. Who cares how you get more profit as long as you get it.
Lastly, and probably the biggest limitation is that the RSI does not equally apply to all periods there are some cases when either there is just not enough data or the frequency of data isn’t enough. In my experience you should at least have twelve periods and you should use no more than a period tracked in months. This is because, as a rule of thumb, I like to think of the RSI period number as a reference for how many periods you can look ahead, and predict comfortably. For example, in the example above the RSI (3) is a 3-month RSI so I would only look to predict the trend in Gross Profit for the next three months. If I used quarters in stead it is like saying that the RSI-3 can give you a reasonable outlook on the behavior of a process over the course of 3 quarters which is nuts. So, as a disclaimer of sorts, I would like to state that I do not recommend using any period value in excess of 1 month. But, also note when I say the three period RSI allows you to predict the next three periods, I mean you will be able to predict the result of the 3-periods not all the movements in between. Which is why, strategic decisions, with their large relative capital expenditure requirements would be better in the case of months or even quarters than stock price as most don’t want to hold on for something that long before changing the tactical exposure of a portfolio.
What does this mean for your business? – Hypothesizing Strategic Decisions based on the RSI
First I would like to say, if you have made it this far, thanks. It is great to know when your ideas are heard. OK anyway, well as far as a helpful tool for you consider this – when the RSI crosses below the 30 in stock analysis it is confirmation of a bearish trend in stock price meaning, it could be a good buying opportunity, just as the probability that a stock will remain in overbought territory forever is unlikely. I guess you could just say pay attention to the phrasing. Above the 70 is OVER-bought while below the 30 is OVER-sold. Now, where it may seem that being over the 70 is a great thing, which it can be, I would look to see how far the next downturn will be compared to the present situation and use the profit to either invest in a product to be released sooner or look to divest in the product line. Basically, consider taking some profit and utilizing it. If instead, the RSI is below the thirty, this may be a good time to “double down” so to speak. Especially, if it is a unique circumstance.
Whelp, that’s it. I hope this works out for some of you. If you have any thoughts feel free to leave them in the comment section and I will answer as soon as I can. Just remember my disclaimer at the bottom. If you are going to use the idea please cite my work. Thanks.
P.S. If you would like the spreadsheets I compiled, I am more than willing to hand them out to people curious enough to check them out.In case you have created this image please get in touch with me through comments and I will be happy to attribute your work. I always make an attempt to identify the owner and ask permission before publishing. All trademarks, copyright or copyleft based material, official and unofficial logos are the property of their respective owners. If you have created or own any particular piece of work which has not been attributed then please get in touch with me at the earliest. All articles, quotes and photos featured on this blog unless or otherwise attributed belong to the author of this blog.