If you are like me, and have been drawn to the world of marketing in part because you are more intuitive than mathematical, you’ll roll your eyes about linear optimization equations. This is an algebra challenge, based on variables, constants and constraints, and has relevance to marketers I think in (a) setting Go/No Go decision parameters, and (b) allocating marketing budgets based on different media/expense options.
The classic application of this stuff occurred in World War II, when military logististians sought to determine the best strategies for given circumstances. One, for example: What would be the most cost-effective diet for soldiers, given specific nutritional and energy requirements and constraints?
First, here (from Google) is a simple example of linear optimization variables:
Maximize 3x + 4y subject to the following constraints:
x + 2y ≤ 14 3x – y ≥ 0 x – y ≤ 2
Don’t worry, if this stuff is arcane to you. In fact, linear optimization calculations can be cumbersomely complex for mere mortals.
The military diet question, for example, defined as the Stigler Diet, related this question in 1943:
For a moderately active man weighing 154 pounds, how much of each of 77 foods should be eaten on a daily basis so that the man’s intake of nine nutrients will be at least equal to the recommended dietary allowances (RDAs) suggested by the National Research Council in 1943, with the cost of the diet being minimal?
Squads of clerks working with calculators worked for days to plug in all the variables, test them, and come up with the answer: A rather unappetizing meal selection, at least for me, since its key component is one food that makes me sick to even think about — liver. (And researchers couldn’t solve the problem to achieve this answer until after the war ended, Google’s research reveals.)
- Enriched wheat flour
- Cabbage beans
These sorts of mathematical problems, of course, are perfect for computer scientists. More recently, Google took the linear optimization calculations and built a tool/application, freely available and inelegantly named “glop”, within its spreadsheet tools.
Creativity, of course, can still be achieved within the implementations. Google chef Anthony Marco designed a plate with the optimized food selections and made it look somewhat appetizing (but I’ll never eat it, I’m afraid.)
How can you apply this stuff for marketing challenges?
Here’s an article, which may provide clues: Kamil Bortacha writes in Marketing Option with Linear Programming:
CMOs need to make complex decisions about budget allocation and marketing investment. Deciding which campaigns will receive funding is never easy, especially with multiple factors and obligations that need to be taken into account. Agency and content preparation costs might be pre-set and an overall marketing strategy might require a certain share of voice or presence. At the same time, customers interact with multiple channels and not all overlapping marketing efforts will be fully incremental.
This can be framed as:
Follow a simple 4-step process:
- Choose the unknowns. Usually, these will be investments or resource shares.
- Write the objective function. What are you trying to maximize/minimize? Costs, margin, incremental value, time spent, reach etc.
- Define all constraints in plain words and then transform them into a set of inequalities.
- Plugin the data into your software of choice and solve.
With Glop, you can get the number crunching done within minutes. Then, maybe you’ll discover a construction marketing option more palatable to your taste than liver, beans and cabbage.