Three Ways to Implement Generative AI

Generative AI (genAI) is becoming the zeitgeist of our time. People are discussing it at home, at the park, in the boardroom, in school, and on Zoom. Companies and executives understand the need to explore these new capabilities, but stumble on a key question: How do I move forward with something I know so little about?” 

But here’s the thing. Your organization already uses genAI. Do people in your company Google? Has your company hired someone in the past six months using online hiring technology or software? Does your organization have Microsoft? If so, your employees are already using genAI tools.

The paradox

There is a common trend among companies adopting genAI-powered tools but stumbling on the adoption of a genAI strategy. It comes down to a single word: value.” People can broadly understand the value of a tool. We understand the value of using a shovel as opposed to our bare hands to build a sand castle or the value of Excel compared to a pen and paper. However, we struggle to understand the value of a concept. What is the value of a lever or the value of math? These are incredibly hard, if not impossible, questions to answer. 

All too often, the value of AI is communicated as a concept. Such-and-such tool unlocks the power of genAI” or powered by genAI” are two examples of where understanding the value of a concept is introduced. 

Organizations should forget about trying to understand the value of the genAI concept and instead focus on understanding the value of the outcome.

When something as transformative as genAI comes to light, businesses need to know where to focus and how to evaluate value, especially when inaction comes with great risk. There are three strategies an organization can choose from when it comes to implementing genAI inside their organization: 

  1. Employee-Led Strategy
  2. Product and Process-Led Strategy
  3. Information-Led Strategy

1. Employee-Led

One approach to genAI is to encourage and enable employees to select the best tools available. Decentralizing the AI strategy by empowering employees to adopt new technologies that suit their needs inherently supports a de-facto AI strategy. By then mandating that employees periodically review their chosen toolkits and claiming a company-wide preference on automation, the business then institutionalizes identifying and adopting genAI tools. 

In an employee-led AI strategy, each individual and team selects the best strategy for their use case. What could be better than the marketing team updating their tech stack to include automated image resizing and social copy creation if that’s what would best serve them? Or the customer success team selecting a tool that automates 80% of calls if that would help them meet their goals? 

Despite the appeal of enabling change and regular review, there is one clear downside: Most blanket invitations to innovate do not work. 

Employees need clear ownership and incentives for driving change. Broad strategic mandates are too easily overlooked. In short, employee mandates suffer from scalability issues, and for large organizations, this approach is comparable to surrendering the strategic vision. 

This is not to say that an employee-led AI strategy can never work. With proper direction and communication, this strategy can be a quick and effective way to upgrade business tools and processes. 

2. Product and Process-Led

The benefit of a product and process-led strategy is clarity. This strategy is targeted and focused. It has clear, limited stakeholders who understand the feature or process being introduced. These stakeholders can quickly identify what is relevant, what the current pain points are, and opportunities for improvement. 

Consider a Software-as-a-Service (SaaS) company. GenAI fundamentally alters how customers interact with products, reporting, customer service, sales, and marketing. Think automated in-app summaries, design automation, e‑commerce product recommendations, and inbound lead communication automation. GenAI’s customization capabilities result in happy customers who expect the same level of interaction across multiple products. Organizations that produce unique products or have home-grown secret sauce” processes and tools need to evaluate upgrading with genAI to ensure those customers are feeling the same satisfaction. 

The two big hurdles facing product and process-led strategies:

  • Education
  • Prioritization

Education

Education is not the thing that stops progress of a product and process-led strategy. It is actually the understanding of value that is the real roadblock. Start with understanding the value genAI can provide, then address any educational need. 

Prioritization

Prioritization can go by many names: roadmap, backlog, resources, commitments…And the objection is often the same: there is a tactical reason why the business cannot be strategic. As Sun Tzu said, Tactics without strategy is the noise before defeat.” In today’s world where yesterday’s technology can be displaced over a weekend, it is paramount to re-evaluate the strategic roadmap and realign the tactical execution.

Product and process-led strategies need to factor in the time and resources needed to understand the value of the genAI opportunity. As part of this evaluation, organizations should revisit the roadmap and understand if new capabilities invalidate the plan. Ultimately, the well-defined nature of a product and process strategy makes it the strategy most likely to succeed, although the weight of existing commitments and priorities presents a significant challenge. 

3. Information-Led

Businesses run on information, and information is dependent on language. For example:

  • Conversations between sales teams and customers
  • Automated system updates, documentation, and presentations
  • Product planning meetings and annual reviews
  • Notes created by machine operators and customer service representatives
  • Product descriptions, ad content, and supply chain management

This language is a source information which relies upon technology and entire teams of analysts to organize and structure it. These investments have been made in order for businesses to gain insights that improve results. Behind the scenes, though, these systems and teams have undoubtedly ignored unstructured information in that — there’s just so much of it! So while an information-led strategy can provide an overview of exactly how a proposed change might impact a business, will there be enough team members to identify the data, context, research, as well as the human expertise to evaluate the outcomes and make a decision?

People are certainly needed to identify the problems and desired outcomes, but this strategy favors well-formed, easy-to-evaluate data. Faced with the choice between quick insight and uncertainty, analysts all over the world would choose the quicker option — getting started with genAI. Such a powerful resource empowers employees to make better decisions, more often, and in a fraction of the time. So while an information-led strategy might work for some, there’s the risk of delayed start. Spending time and resources further evaluating genAI’s future role in your business case could put an organization behind their competitors who embraced the change more quickly.

Get Started

The applications for genAI are endless: content creation, document summarization, billing, ticketing, call centers, service requests, business analytics, product descriptions. It’s time for organizations to be proactive about unlocking new capabilities for structuring, managing, and gaining insight into any process that involves language. All that’s left is to pick a strategy and in some cases, a partner, to implement genAI and begin. 🚀