Engaging consumers through conversations: a management framework to get it right across converged channels

Paul Pangaro, Hagen Wenzek


Paul Pangaro is CEO of General Cybernetics, Inc., applying conversational models to content publishing, enterprise collaboration and education. His career spans product strategy, prescriptive innovation and organizational dynamics, with roles as technology executive, professor, entrepreneur and performer. He spent ten years in Silicon Valley as CTO and strategy consultant to startups, including a stint as Distinguished Market Strategist at Sun Microsystems. He taught the cybernetics of design at Stanford University, currently teaches at School of the Visual Arts in the Interaction Design MFA program, and has lectured in São Paulo, Paris, Berlin, Vienna, Toronto, and in cities in the US. He holds a doctorate degree in cybernetics and an undergraduate degree in computer science/humanities. Email: pangaro@generalcybernetics.net Twitter: @paulpangaro

Hagen Wenzek is the principal and founder of Freestyle Consulting, an advisory business that helps companies navigate the digital ecosystem. His experience as the Chief Technology Officer of media holding company IPG Mediabrands, at IBMs Corporate Strategy, and with IBM Consulting led him to the intersection of technology, media, marketing, and consulting services that forms the center of gravity for widespread innovation. Start-ups in digital marketing and big data look for his advisory on how to become a trusted business and mature corporations leverage his consultancy to work with these innovators. He holds a doctorate degree in electrical engineering and a number of advisory board positions with those very start-ups. Email: hagen@freestyle-co.com; Twitter: @hagenwww



Marketing performance is heavily dependent on the level of consumer engagement with a product or brand. Delivering a brand message has traditionally been a major means to engaging consumers, but the messaging was limited by the nature of a one-to-many communication that only allowed for very limited feedback from the consumer. With the advent of digital convergence, the transition from communication to conversation becomes not only economical for the brand, but is expected by the consumer.

However, to effectively and efficiently manage a conversation through all the traditional and new media channels with many – if not most – of a brand’s consumers requires a rigorous approach and comprehensive methodology. We are presenting a framework that helps a brand or their agency to understand and actively manage all the critical components of a conversation. Following the CLEAT-Framework – Context, Language, Exchange, Agreement, Transaction – sets up a conversation for success and allows a measurement of progress along each stage, from understanding the situation of the consumer all the way to (ideally) a commercial transaction. With or without an immediate commercial transition, engaging consumers in true conversations is the most effective way to build persistent relationships of trust and loyalty – the goal of high performing marketing.


Keywords: Conversations, consumer engagement, marketing effectiveness, marketing performance, digital convergence

Marketing Performance

The desired effect of marketing is to drive sales of a product or service while commanding a higher price, as justified by the consumer valuing this brand more highly than that of a competitor. The means for doing so is to reach the consumer with the right message at the right time. To send this message, the marketer can use a growing number of media channels.

The choice of channels, messages and their timing determines how effective marketing is. Determining overall performance, however, also requires factoring in the efficiency of that process. For the longest time the impact of improving efficiency was small compared to the cost of using the channel, i.e., paying the price of advertisement inventory; therefore, optimizing efficiency did not garner much attention. This shift in importance is reflected in the development of dedicated media agencies that manage the process of delivering a brand message on behalf of a brand owner:

Until the late 2000’s media buying was part of the creative activity of designing the brand message and had rarely been managed separately. Then media buying was carved out and consolidated across many accounts to leverage supreme negotiation power and decrease the price for media. Only after the media buying process was separated from the creative process could each of them be discreetly optimized., This optimization focused on effectiveness of marketing tactics, e.g., by targeting specific audiences rather than mixing demographies to improve outcome. Now, even further advancements are being made in the improvement of efficiency in managing, buying, placing and interacting with media and messages, exemplified in the acceleration of programmatic trading.

Digital Convergence

Back in 1978 Nicholas Negroponte saw the emergence of Digital Convergence [Brand, S. 1987] [i] as the coming together of all compute, print and broadcast media. Later it became clear that it is not just the mere “coming together”, but the “takeover of all forms of media by one technology” [Mueller, M. 1999] [ii] — the underpinning of all content, communication and computing by digitization in general.

Before digital convergence, each medium that transported a message could be managed separately and the end-user experience was highly controlled around a captive consumer. Thus brand messages were tied together only by a common brand strategy, while the execution happened independently for both creation and transport – and therefore expensively. With a converged marketing channel mix, that situation changed profoundly. What is built once can now be used for multiple channels, yielding efficiency gains through reuse but also increased effectiveness that emerges from a consistent experience for the consumer. That is complemented by metrics that describe performance consistently across all consumer interactions on all channels

The disruptive nature of that transformation makes it self-reinforcing. Cost reduction is inherent in digital channels and releases funds for technology that further drives digital convergence. Taking all these developments into accounts led us to adapt the very definition of digital convergence:

§  Content is not tied to a specific format anymore, but becomes independent and easily transformed from one format into another.

§  Communication is not delivered from a central instance anymore, and control moves to the edge of the network where the consumer drives interactions that may happen at any time, on any device, and in any direction.

§  Computing vanished from servers owned or controlled by the brand and becomes a service delivered from the cloud.

Therefore digital convergence describes the decoupling of content, communications, and computing from the constraints of physical media, devices, and machines through digitization.

Consumer Expectations on Conversations

As digital convergence has a large impact on marketing performance; it also raises the expectations of the consumer towards interaction with the brand. The ultimate digitally converged media scenario is social networking. It is not only a media channel itself, it is a multidirectional communications medium that relies on personalized content accessed over multiple devices and leverages vast cloud computing resources. But more than anything, it is allows a brand to draw a consumer into a conversation.

A conversation is an ongoing engagement in which the participants receive a benefit, where the benefit may be in the realm of new or changed ideas, understandings, beliefs, or intentions, which in turn affect further actions and transactions. [Dubberly, H., and Pangaro, P. 2009] [iii] Since a rigid definition is absent from todays technology and media world, the following sections present a breakdown of the actionable elements as well as quantifiable metrics of a conversation. In short, however, learning how best to foster conversations to meet and exceed consumer expectations leads to higher engagement that, as we will show below, leads to greater marketing effectiveness.

The vast possibilities opened by conversational engagement come with considerable responsibilities and it is therefore mandatory to understand what having “an efficient and effective conversation” [Blair, G. 1992] [iv] really means:

§  You must make your message understood.

§  You must receive/understand the intended message sent to you.

§  You should exert some control over the flow of the communication.

From a brand management perspective the responsibility for having the recipient understand the message and governing the conversation lies fully on the side of the brand. It is the expectation of the consumer that the brand will leverage all the converged channels available to that consumer. In the example of social networking this would mean using the same “hashtag” on each site and channel; moving a conversation forward that might originate on TV or print through interaction on Facebook; or, transitioning seamlessly from Twitter to the owned media website to fulfill an aspired transaction.


All aspiration of a brand to achieve marketing effectiveness hinges on a high level of consumer engagement. While consumer satisfaction was historically seen as a key indicator for a strong brand relationship, recent research [McEwen, 2004] [v] shows that it is far better that “engagement […] be used as a proxy measure of the strength of a company’s customer relationships based on the extent to which customers have formed both emotional and rational bonds with a brand”.

That new insight led to follow-up research going deeper into the differences between new and existing customers. Before digital convergence, that distinction was of primary interest in direct marketing, because only there could the message be targeted and tailored to the recipient with sufficient precision. Now, however, brand messaging through digital and social channels can easily take very specific information about the consumer into account. Therefore, understanding different types of engagement is not just relevant, but becomes critical for marketing effectiveness.

The main aspect for new consumers to evaluate a brand’s product or service is how much they rely “on attribute-level information” [Patterson, P. 2000] [vi] and “on tangible and often extrinsic cues” [McGill, A. and Iacobucci, D. 1992] [vii] which can be exemplified by the “generation of comparison standards”. This would very much explain the success of digital media sites and special sections dedicated to reviews, such as Amazon.com or Yelp.

Existing consumers often do the opposite of what is observed for new ones and “often blend negative statements out and emphasize positive ones” [McGill, A. and Iacobucci, D. 1992], thus act much less on the information that others provide about the brand, but on a conversation with the brand itself.

Measuring the difference in engagement via, e.g., a Link-to-Conversion-Rate on Facebook shows the dramatic difference. A study in collaboration with UNIFIED Social, a social media service provider, of 2,500 unique Facebook ads (see tables 1 and 2 below), showed that existing consumers (proxied as ‘fans’, see Table 1) were more than seven times more likely to follow through a conversation than new users. The incremental cost for these conversions, i.e., one further step in a conversation, was also 75% lower (see table 2). Therefore one can deduce that an engaged consumer can be moved along the marketing worfklow much more successfully and at lower cost. So marketing effectiveness – the measure for how well money is spent to achieve a marketing goal – goes up.

Fig 1

Table 1: Brand Strategies and Targeting Characteristics [Friedman, D., 2013] [viii]

Fig 2

Table 2: Brand Strategies and Targeting Characteristics [ibidem]


As we can now safely assume that marketing effectiveness is positively correlated with customer engagement and a conversation is a means to increase and sustain engagement, the question is how to manage conversations between brands and consumers.

Looking at the background of digital convergence and marketing effectiveness makes it clear that managing conversations is a very complex activity:

§  Engaging a consumer is key to make marketing more effective.

§  Conversations need to occur across all digitally converged media channels to be comprehensive.

§  Conversations need to be highly tailored to the customer, domain and every stakeholder’s expectation.

Coping with that complexity is best done by leveraging a framework that lets the expert focus on distinct elements. In management consulting, frameworks are broadly used to identify and break down problems as well as provide guidance towards finding a solution. Thus a framework is not a detailed procedure to be followed, but a guidebook and needs to be used as such.

Our framework provides insights into five distinct components that, when instrumented effectively, will lead to successful conversations, specifically when the goal is to increase the number of commercial transactions through effective marketing. These components are: context, language, exchange, agreement and transaction.


Traditional mass media communication takes little contextual information about the audience into account, because it is not conversation but just the delivery of a message that has been the goal. However, digitization of media enables (big) data and intelligent algorithms running on cheap, scalable (cloud) computing infrastructure to harvest that very data and to derive insights about the consumer’s context. That context can then be used to make the conversation highly relevant. Seeking to answer the following questions helps to decide what data to analyze and how:

§  Whom to talk to? What is the right audience?

§  When to talk to? What is a good moment?

§  How to talk? What is the best channel?

§  What to talk about? What is the best content?


Selecting the right audience is a domain of extensive research in itself. Traditionally in paid media, the audience reached via advertisement is based on the demographic segments selected, e.g. ‘females 18-35’. In digital media, targeting can become much more refined because the attributes that are known about individuals accessible on a digital channel are exposed to advertisers in groups or even individually, e.g., by retargeting based on previous digital activities (via ‘cookies’) or alignment with email address lists from customer databases. Most sophisticated service providers can offer audiences based on previous results of campaigns from other advertisers, which can reveal unexpected audiences with whom to start a conversation.

Another angle for selecting an audience is to take the viewpoint of profitability. Here, building an audience profile that correlates with the characteristics of a predicted high Customer Lifetime Value (CLV) would align not just with marketing goals, but overall business strategy. (For CLV characteristics see e.g. [Reinartz, W. and Kumar, V. 2003] [ix] and for a system to calculate a Social Lifetime Value e.g. [Katana] [x] ).


The right timing for sending a message to initiate or move a conversation forward is a critical but often mismanaged aspect of context. Using detailed data about campaign interactions between brands and consumers, combined with the intrinsic aspects of the channel, creates actionable insights about the right timing that were just guesswork before. Observing those interactions lets us apply more scientific rigor to data gathering and analysis especially on social networks. For example, by looking at the level of engagement that a promoted story on Facebook achieved during different time windows after an event, a dramatic drop is seen for periods longer than 24 hours [Unified 2013] [xi] . This is a combined result of how Facebook ad serving algorithms place the story as well as consumer expectation. As other channels and networks have different characteristics, running experiments to understand them individually will be critical.

Channel & Content

Content itself will be discussed under the ‘language’ component of the CLEAT-framework below. However, contextual information about other content relevant to the conversation partner across multiple channels is key to selecting the right content. Usually, the context of a consumer engaging on a single channel is analyzed through focus group studies and extrapolated to a larger demography. However, under the umbrella of “companion devices” or better “companion channels”, content delivered via one channel, such as TV, is complemented by activity over another one, such as Twitter on a mobile device. This companion-activity offers the unique opportunity to understand the context of an individual consumer’s engagement and thus to tailor the message much more precisely to drive the conversation.

Services that provide correlated analysis have become available from established media metric companies such as the Nielsen Twitter TV Ratings [xii] .

Leveraging Natural Language Processing (NLP) decreases the effort to identify relevant conversations, as one is not limited to finding specific terms, but software automatically extends the search space to terms with the same meaning (e.g., see [Morrison, A. and Hamby, S., 2012] [xiii] ).


A shared language is an obvious requirement for any meaningful conversation. However, as one broadly quoted statement points out:

“It is a common misperception that language use has primarily to do with words and what they mean. It doesn't. It has primarily to do with people and what they mean. It is essentially about the speakers' intentions.” [Clark, H. and Schober, M., 1991] [xiv]

Therefore, choosing the right language for each step in a conversation requires much more thoughtful decisions than just selecting whether to send a message in English or in native tongue. Any decision about how to encode and decode information is a decision about language [Wikipedia: Language] and each of these decisions indicates the intent of the brand as well as that of the consumer. For the brand, that intent needs to be aligned with the expectation it has for the conversation (which is captured in the agreements and transactions it wants to achieve as discussed below) as well as the consumer’s intention as it might be exposed through the context (s.a.).

Language decisions are therefore manifold and span communication systems as broad as:

§  Natural Human Language (English, German, Chinese, etc.)

§  Technology platform (Facebook, Twitter, brand website, etc.)

§  Content types (video, imagery, sound, etc.)

§  Content forms (casual, formal, ‘native’, etc.)

§  Vocabulary (generic or ‘insider’)

§  Jargon and abbreviations (technical, social, etc.)

§  Level of abstraction (about ‘why’ vs. ‘how’; reasons for acting vs. possible actions; strategic vs. tactical)

§  Human appeal and values (rational or emotional; logical or visceral; transparent or tacit)

§  Required pre-cursor concepts (assuming the consumer knows they have a need vs. articulating the need)

As it is clear from the scope and depth of these characteristics, the component of language in conversation is complex indeed. But not to address it effectively is to squander media budgets, digital or not. Conversely, when a common language is found – or evolved over time between a brand and a consumer – communication can happen efficiently, because all parties can understand the intent of the conversation and with less overhead over time. For example, the simple press of the ‘Like’ button on Facebook is universal across all more than 1 billion users to indicate approval of and/or agreement with a piece of content. The same is true for a ‘Dislike’ on Youtube that actively signals a lack of agreement and indicates the intent to see less of these type of videos (or video ads).

The recent development of ‘native advertisement’, or ads embedded into the content stream in a seamless way, strives to keep the attention of the consumer on the ad by leveraging the same language that is used to present a connected story.


Exchanging ideas and beliefs in an evolving stream is the core process of any conversation. However, if it is not evolutionary, i.e., if the conversation does not move forward in a positive way, it will stall or be terminated. Furthermore, the term “exchange” also implies that each give-and-take adds value to each party. That value is obvious when the content itself is appreciated, like a good answer to a question (which is brilliantly executed in one of the most effective companies in the world, Google, whose search results are precisely these answers). However, observations about the conversations can also create new insights into behavior, expectations, level of trust, predictions of future reactions, etc,. Those insights might be much more valuable in aggregate than each individual conversations – if they can be harvested.

For the brand, the increase in Customer Lifetime Value might be used as a quantitative metric of the value of an ongoing conversation (assuming the implementation of CLV at the brand level includes behavioral components). The CLV metric can be understood as the monetization of engagement, which stands in direct relation to more granular metrics such as type and frequency of exchange. All these metrics can be used to determine further investment potential to lead the conversation, e.g., through paid media.


Each exchange leads to an agreement or disagreement on the value of the conversation and thus each exchange is critical in determining whether the next exchange will take place, that is, whether the conversation will continue. A conversation is fragile in that either side may cut it off for any reason, although it must be pointed out that conversations that appear to terminate between consumer and brand inevitably continue between consumer and consumer, spreading ill-will and negativity with potentially large consequences (see e.g. [Evans, D. 2012] [xv] ). Complicating that fragile relationship is the fact that algorithms (e.g., on Facebook) might prevent the consumer from ever seeing the next exchange by the brand and might assume disagreement. Using the most recent tools and services to manage system behavior (e.g., via promoted stories) is therefore a crucial competence for leading conversations.

A shared understanding of the value of the conversation increases the level of trust between the parties and therefore leads further along the path to the ultimate purpose of the conversation. For the brand, that purpose has to be predetermined before any conversation is initiated [Blair, G. 1992] and at the end “a clear understanding of the outcome” has to be given [ibidem].


That very outcome in the sense of the brand interaction is a commercial transaction where the consumer buys a product or service. This has the most direct impact on CLV through a monetary increase of past transactions. However, as CLV emphases value over the lifetime of a consumer, the brand should use the opportunity of an existing conversation, even when it concluded with the transaction, to re-engage using the very same framework and make the same decisions about context and language. Just that this time, the data collected about the previous conversation will inform the following one to increase efficiency and effectiveness.

Success Metrics

‘What doesn’t get measured, doesn’t get done’. As this truth is pervasive across companies, defining metrics that quantify results of actions taken to engage in conversations is crucial to manage brand effectiveness.

In the following table we summarized key metrics for each stage of the CLEAT-framework. We have incorporate metrics that (a) look directly at each action a consumer is expected to take, (b) can be collected to represent the outcome of actions during each stage and (c) support decisions for moving each stage forward by looking at the larger picture of, e.g., an audience or group of existing customers rather than individuals.


Fig 3  


Presenting these metrics in an ongoing basis to ‘conversation managers’ in a format that is compatible with that marketer’s expectations is critical for the success of implementing the CLEAT framework. Social media dashboards, CMO-desks, etc., are straightforward places to include these conversation metrics. Those business intelligence systems are actionable visualizations of big data that are being embedded into marketing departments and marketing service provider workflows.


Social Media is repeatedly being named one of the big themes driving any business these days. Conversations provide a purpose to social media as they directly correlate to brand engagement and consequently marketing effectiveness. Marketers and marketing professionals at agencies cannot manage large numbers of conversations effectively, let alone efficiently, without supporting tools. As a management framework CLEAT provides a handle on the key components that make those conversations successful.

With the increased use of analytics and automation to execute campaigns across all channels also comes more and more process performance data. That will help supplementing the CLEAT framework with more precise KPIs, tools and techniques to further improve marketing effectiveness at lower and lower overall cost.




[i] Brand, Stewart. The Media Lab: Inventing the Future at MIT. Viking, 1987.

[ii] Mueller, Milton. "Digital convergence and its consequences." The Public 6.3 (1999): 11-28.

[iii] Dubberly, Hugh, and Pangaro, P., “What is conversation? How can we design for effective conversation?”, Interactions Magazine XVI.4 — July / August (2009): 22-28. http://www.dubberly.com/articles/what-is-conversation.html.

[iv] Blair, Gerard M. "Conversation as communication." Engineering Management Journal 2.6 (1992): 265-270.

[v] McEwen, William J. “Why Satisfaction Isn’t Satisfying,” November 1, 2004. http://businessjournal.gallup.com/content/14023/Why-Satisfaction-%20Isnt-Satisfying.aspx.

[vi] Patterson, Paul G. "A contingency approach to modeling satisfaction with management consulting services." Journal of Service Research 3.2 (2000): 138-153.

[vii] McGill, Ann L., and Dawn Iacobucci. “The Role of Post-Experience Comparison Standards in the Evaluation of Unfamiliar Services by Ann L. McGill and Dawn Iacobucci.” Accessed September 11, 2013. http://www.acrwebsite.org/search/view-conference-proceedings.aspx?Id=7358.

[viii] Dylan, F., UNIFIED, Offsite Conversion Performance between Different Targeting Strategies, New York, NY, November 2013 (internal)

[ix] Werner J. Reinartz, V. Kumar (2000) On the Profitability of Long-Life Customers in a Noncontractual Setting: An Empirical Investigation and Implications for Marketing. Journal of Marketing: October 2000, Vol. 64, No. 4, pp. 17-35

[x] Katana Social Analytics Engine by NinjaMetrics, http://www.ninjametrics.com/social-value

[xi] UNIFIED, “Smart Social Strategies. The 24 hour rule”. New York, NY, 2013 (internal)

[xii] http://www.socialguide.com/nielsen-twitter-tv-ratings, accessed November 6, 2013

[xiii] Morrison, Alan, and Steve Hamby. “Natural Language Processing and Social Media Intelligence.” PwC. Accessed September 11, 2013. http://www.pwc.com/us/en/technology-forecast/2012/issue1/features/feature-mining-social-media-intelligence.jhtml.

[xiv] Clark, H.H., & Schober, M.F. (1991). Asking questions and influencing answers.  In J.M. Tanur (Ed.),Questions about questions:  Inquiries into the cognitive bases of surveys (pp. 15-48).  New York:  Russell Sage Foundation

[xv] Evans, Dave. Social media marketing: An hour a day. Wiley.com, 2012.