Marketers find themselves with a wealth of options in marketing technology today. The number and the range of tools available exceeds all previous options. 

But with this wealth comes a related analysis paralysis. How can marketers pick among the many alternatives to ensure sustainability, flexibility and efficiency?

David Raab and I continued our discussion, looking at the hard choices marketers face when selecting marketing software.

David Raab: Let’s assume that many systems meet the criteria of cloud-based and service oriented architecture with open API. Marketers still need to choose which ones to buy.   

To organize those choices, I think we need to design a true “stack,” in the technical sense of functions divided into distinct layers with structured connections between each layer, allowing us to replace components within any layer without disturbing the other layers. 

My guess is the design of such a stack will be much more stable over time than a traditional “roadmap.” But once the stack is designed, we still need to consider actual user requirements when selecting specific components. It’s true that those requirements are continually changing, but that doesn’t mean we can ignore them altogether.

Mayur Gupta: The challenge is not only in the designing of the stack but first in the assembling of the stack and then in the application of the stack. 

Thinking about user requirements from a marketing technologist standpoint should no longer be about features and functions. That is now for the marketing technologist to figure out and translate. 

The requirements should be business and consumer driven, for instance:

Business Requirements 

  • Increase Market Share 
  • Maximize Lifetime Value 
  • Increase share of wallet 
  • Drive household penetration 

Behavioral Needs

  • Drive Trial 
  • Increase Loyalty 
  • Build Trust 
  • Or resolve for issues like Stigma as a behavioral problem, related to adoption and usage for certain type of brands  

The real opportunity — and at the same time the challenge — is can we design a stack that will address these needs instead of saying, we need to deliver “personalized and contextual” experiences? Phrases like that mean nothing if you’re responsible for running and growing the business. 

Secondly, how do you then apply the stack to solve for these issues? How do you use marketing automation, personalized content strategy and a channel-agnostic experience planning framework to ultimately drive and maximize lifetime value? 

That’s the opportunity and the requirement. 

DR: So we find ourselves in a situation where we 

  1. design a stack in terms of functional layers 
  2. specify technical criteria that stack components must meet, at least in terms of how they communicate with other layers 
  3. identify components that match those criteria and 
  4. choose the qualified components that best match current and near-term future requirements for driving long-term value.  

That sounds reasonable in theory, but still leaves some hard questions to resolve. 

For starters, how do you know what stack design is right, especially in terms of future flexibility?  And, if it later turns out that your chosen path leads to a dead end, how do you recognize that and change direction?

MG: You don’t know and you won’t know. 

That’s why marketing technologists have to adopt a big testing and big learning mindset — whether optimizing a campaign or optimizing your technology stack. Operate in short sprints, run pilots, proof of concepts and then scale. 

How do you change directions? That’s where SaaS and cloud based solutions, usage-based pricing and loosely coupled architectures come in. 

Otherwise, using the traditional thinking of home grown solutions, by the time you built the perfect solution for your set of requirements, the world and the consumer needs would have evolved dramatically. The need for speed is far more powerful in today’s consumer-led world than the desire for perfection.  

DR: How do you embed data flexibility in your stack design?  Future marketers will need to store ever-large volumes of data and deal with a stream of new formats and structures.  

Service-oriented architectures (SOA) and open APIs don’t really address this.  Nor is it clear that you can isolate the data issues within a single layer of the stack.  

MG: Very true. However, a few things happened in the last three to five years that opened many doors from a data convergence and harmonization standpoint. 

So while APIs and SOA don’t solve for data storage, they do solve for data visibility and integration. With an API-driven ecosystem, your data can flow from one end to the other, from one system to the other system, tying the consumer’s journey together.

However, that still may not solve the fundamental issue around data fragmentation. This is where the new breed of DMPs (Data Management Platforms) play a massive role. 

They’re doing the brunt of the data collection work across first party, second party and third party, and doing data matching and data mastering both at a consumer and household level. For the first time these DMPS are truly giving a 360 view of the consumer. 

I see the APIs as the PVC pipes that enable the data to flow throughout the machinery and the DMPs as the filter that soaks in data from all sources, masters it and makes it available to the marketer. 

You still have to run the final yard to transform the data into insights and action, but now you have raw gold. 

DR: This all works great until the only system that meets marketers’ functional requirements doesn’t match the marketing technologists’ flexibility criteria.  

Telling marketers they can’t have what they want sounds an awful lot like the bad old days when IT tried to dictate to marketers, which is one reason we have marketing technologists in the first place. 

MG: It’s a difficult decision, but it happens all the time. I go back to the point about the need for speed vs desire for perfection. 

What I have done in the past is evaluate the ROI and the impact to the business. Is the ROI strong enough to go with the platform despite the limitations of flexibility and extensibility, even if it’s for a short period of time, say two to three years? Does it help you pilot a new capabilit or business model while you figure out a long-term strategy and solution that is scalable across the globe, across multiple brands? 

It all boils down to the “problem you are trying to solve” but is by no means is a show-stopper. 

DR: I know I’m still dealing here with the technology level, not even addressing your larger questions of becoming customer-obsessed and ROI focused. But can we really shift to that level before solving the technical questions? Or do the solutions to the technical questions simply have to take those larger issues into account?

MG: What I’ve learned over the course of many years is that the best technology solutions begin with an understanding of the consumer behaviors that need to change and the key business problems they need to solve. 

More often that not, we get technology obsessed for two reasons: 1) it is exciting and 2) it is easier to think and start like that. 

technology obsessed

But it’s often hard to get out of that mindset and really put the consumer at the center. That’s when we have great technologies and tools that don’t often solve the consumer need or change behaviors. 

I see a pattern right now within the world of marketing technology: the focus is too much on the volume and expansion of the landscape, and with this comes a rush to bring in and adopt a high number of these capabilities. 

But where is the measurement of the ROI from this investment? Most stats and figures from analysts today focus on the investments being made in this category, the number of acquisitions, PE capital etc., etc. but we have not yet reached a point where we are measuring the true top line growth at a macro or micro level.  

That’s why I think the journey begins with the consumer and the business need, and then tying that to the technology design and stacks. 

Editor's Note: Read part one of David and Mayur's conversation here

For More Information:

Title image Siobhan Fagan