Modeling the consumer's mind

There has never been a greater amount of information about customers and the market, however, key aspects in the purchase and decision making processes still remain obscure. In a hyper-connected world, consumers are no longer passive recipients of messages but also creators and generators of the same trends. By knowing the emerging dynamics and the way in which contact points with the brand influence these decisions, a respectable degree of control can be achieved.

The old techniques of segmentation and market analysis based on deterministic models are no longer effective by the exponential increase of complexity that characterizes the current market. Marketers now face the difficult task of succeeding with fewer resources to handle increasing thresholds of uncertainty.

With Zio ® ABM we can replicate to scale the market in which our brand has to evolve. In this virtual environment we represent not only our brand, but also competing brands and a population of agents (consumers) whose rules of behaviour we intend to reveal. The agents form a social network by virtue of its connections with other agents and this network respond to stimuli created by the touchtpoints of the brand, the experience of use with the products and all possible forms of communication between brands and the social community.

Using historical data, artificial intelligence algorithms in a regressive simulation are able to reproduce what happened in real life to the present moment. From here, with the model already calibrated, we can project future alternative scenarios and test different ideas like in a social laboratory.



Main uses and applications

Modeling of a population of agents provides an understanding of the rules of consumer behaviour as it happens in real life, individually, per segment and as a whole. Modeling techniques allow to know how agents become influencers for other consumers through word of mouth and how these phenomena interact with marketing actions and product strategies. One of the main uses of the models is to optimize the marketing budget through optimal media mix selection with the highest degree of effectiveness for our audience. It also allows measurement of the effectiveness of ad campaigns and product brand positioning as well as detecting the concepts and attributes that are most relevant to our target audience.

Simulation with models allows us to compare different strategies and choose which are the most effective or appropriate with the reality of the company. The possibility of testing before jumping into reality is the best way to minimize errors and optimize results.

The concept of systemic modeling
with agents provides a dynamic view of the effects that provoke the actions of our competitors, and thus we can run simulations to counteract such actions or evaluate ideas and initiatives prior to going to market.

With ABM we can identify "how" and "when" consumers change their behaviour so that our client’s brand is able to influence this process.
 
Analyze what changes we should make on the marketing strategy to maximize the KPI's defined in the model: sales, ROI, awareness, brand equity, and word of mouth.
 
Identifies which customer segments are more able to influence the decision making process and assesses if I'm acting effectively on them.

Select the messages that are most likely to impact directly on the equity of the brand.

Allows for constant updating of information, thus making the software a true tool for dynamic brand management.



What is the level of accuracy of the models

Thanks to the calibration algorithm that operates on historical data, we increase the confidence in the system to predict future events with accuracy rates close to 95%.


Major advantages over other modeling techniques

We represent the market as a whole and not just some of the participating brands.

Enables an integrated management of paid, owned and earned media on a single platform.

Compared to econometric models that are deterministic, ABM models generate results with ranges of variability to assess the risk associated with each of the analyzed strategies.

Models are based on individual rules and facilitate the incorporation of expert knowledge.

Allows evaluation of the impact of social relations on the model.

Because of the open nature of the models, working hypotheses are explicit.

ABM models are dynamic and they accept continuous changes in real time.

Holistic approach: allowing networked information to understand emerging phenomena. There’s no a single dependent variable and many independent variables, but all model variables are interdependent.