by adgage blog>
An Outdated Approach
A typical marketing cycle consists of two main phases: a planning phase and an execution phase. During the planning phase, the marketing team decides both the financial and creative components for the following period, which may be any length of time, such as year-long or quarter-long. The financial component consists of determining what resources the department will need to execute the marketing plan. Meanwhile, the creative component includes identifying the content, audience segments, channels, and timeframe for the plan. Finally, the execution phase involves implementing the marketing plan as close to schedule — and budget — as possible. This includes adjusting content, plans, and budgets as situations arise or markets change during the time period.
Unfortunately, this approach has become outdated in today’s world of highly varied, high-volume traffic that floods the online world. Real-time marketing plans must be able to react quickly to changing market needs, and media-mix planning brings a new approach to marketing that aims to meet that need.
Media-Mix Planning
Media-mix planning helps to determine the right budget mix to maximize the effectiveness of a marketing program. There are two main approaches to media-mix marketing: top-down and bottom-up.
The top-down approach involves looking at high-level spend and historical revenue data to predict how your plan will perform moving forward. This type of approach is considered easier to execute than the bottoms-up approach, as it looks at higher-level, concrete data as a basis for its predictions.
The bottom-up approach is considered to be more performant, as it provides both high-level and more-detailed insights into strategy. However, it is more difficult to execute this approach based on its inputs. The bottom-up approach looks at more granular cookie-, user-, and creative-level data points for its predictions. These data points are difficult to capture and correlate, especially in offline scenarios. It is also difficult to connect economic data with the bottoms-up approach, making it challenging to accurately assess the potential ROI for your program.
Whatever the approach, media-mix marketing improves both phases of the typical marketing cycle by increasing the accuracy of ROI predictions based on past performances. In turn, this will promote better forecasting and budgeting. It also improves the execution phase by enabling real-time insights into changing market behaviors and providing solutions to capitalize on trends. One of the major components of media-mix planning is the ad-stack.
Figure: Unified Planning and Execution of marketing requires an Ad Stack. Various marketing efforts and their outcomes are captured by the the analytics framework. This is fed into a modeling engine that returns the media mix and flighting plan. A bottoms up framework also return a marketing plan at a cookie/audience level. This is then fed into the execution engine comprised of a Data Management Platform and the media buying engine.
Understanding the advantage of an ad-stack
The ad-stack consists of a mixture of creative, designed to be executed across multiple channels and to capture important, real-time analytics on the behaviors of the people interacting with it. It allows a multitude of content to be managed in a single space — from online e-commerce to mobile sites and apps, across your social outlets, and even through specialty devices such as in-store kiosks. All of these channels can be served from the ad-stack, and the resulting data points can be correlated to provide unique insight into the market.
The ad-stack allows these data points to be captured in real time, which enables faster response times when changing content within a marketing program. A faster reaction to changing markets equals greater conversions on campaigns that, historically, had the potential to waste large amounts of company resources and result in negative brand impact. If real-time markets need real-time analytics, then the real-time analyzation capabilities of the ad-stack allow media-mix planning to meet that need.
Finally, the ad-stack promotes intercompany synergies by helping employees achieve a better understanding of the sales process. More-accurate predictions improve financial resource needs, shocks to inventory and shipping — and most importantly, they better prepare sales for the industries and audiences they are pursuing.
The adoption of media-mix planning presents a change in a business’s approach to its marketing. It signals a turn-away from the historical two-phased planning approach of the past in favor of a more dynamic way of meeting the shifting needs of the market in the future. This approach minimizes the impact of purely hypothesis-driven marketing, while it permits more insight through a trial-and-adjust approach — one that captures millions of data points and automatically adjusts the customer experience based on real-time trends. The opportunities for ROI afforded by automated-machine learning are too great to ignore, and the combination of a media-mix planning approach and the capabilities afforded by a well-designed ad-stack can bring your marketing program into the 21st century.
Sid Shah
Dr. Siddharth “Sid” Shah, director of business analytics for Adobe’s Digital Marketing Business, and his team oversee the performance and strategic development of its client accounts. He has a background in algorithm development and statistics which he formerly used to push biomedical imaging boundaries and now uses it to enhance the performance of his clients’ campaigns. Although Dr. Shah has a penchant for bad nerdy one-liners ("Don’t drink and derive, alcohol and algebra don’t mix"), he is always interested in explaining complex mathematical methods in simple, elegant ways to business audiences. When he is not working with business models (which he finds more beautiful than super-models), he can be found hanging off cliffs in some of California's National parks. Previously, Dr. Shah was director of business analytics at Efficient Frontier, which Adobe acquired in January 2012. He has a double masters and PhD in Biomedical Engineering from the University of Michigan.