Marketing AI Strategist: What To Look For

the ai strategy consultant

The good news is that AI strategy is no longer the exclusive domain of data scientists and developers.

The bad news is that these days it seems like everyone’s touting themselves as a marketing AI strategist. If you’re looking for help with your AI strategy, how do you know who to trust?

Let's be candid - many of us, including myself, were not actively engaged in the AI space prior to the meteoric rise of ChatGPT in November 2022. That in itself should not preclude one from becoming an AI strategist. Before this watershed moment, deep AI/ML capabilities were siloed within tech giants. Therefore, unless one worked at those elite firms, AI expertise was hard to attain.

At the same time, failing to carefully evaluate your consultant will lead to problems, which can range from legal and regulatory risk, PR risk from unanticipated bias, to significant financial losses.

When engaging an AI strategy consultant, don’t jump without looking. Ensure your partner has these key qualifications:

  1. significant domain expertise

For the purposes of illustration, I’m going to use marketing analytics as the example, because that is my area of expertise. You can substitute your own.

The consultant should possess substantial hands-on experience in the domain you’re engaging them for. For example, an AI strategist helping implement marketing analytics solutions should have 10+ years of practical marketing analytics and operations experience.

For example, a company we know engaged a renowned data scientist to build an AI strategy for the marketing team. They didn’t understand how marketing cookies work, and didn’t understand how to present data for marketing decision making. The solution had to be scrapped.

Relevant marketing analytics credentials should include:

  • Training personnel in best practices for marketing analytics and visualization, including “last mile” expertise in getting data into the hands of decision makers

  • Creating governance strategies

  • Presenting data insights persuasively to executives

  • Overcoming resistance to process changes and new insights

  • Delivering practical solutions, not shelfware or vaporware. In particular ask about their experience with pilot projects, and scope your first engagement tightly to a domain that will benefit the most from retraining. For example: being able to move some marketers away from preparing analytics data to developing more creative content and strategies.

Simply having AI experience at marquee tech firms like Google or Meta does not necessarily translate to domain expertise. Beware of influencers without real-world implementation experience.

2. commitment to ethical ai principles

Given the current pitfalls in deploying AI, prioritizing ethics is paramount. Avoid those who cut corners or fail to grasp the regulatory and reputational risks. Verify they share your organization's values around responsible AI through their track record and belief system. Can they articulate best practices around security, governance and compliance? Do they appreciate AI's potential harms if misused?

3. ABILITY TO BUILD AI SPECIFIC BUSINESS CASES AND FINANCIAL MODELS

Even if not building in-house, your consultant should understand the total cost of ownership of AI solutions. Unlike traditional software, AI has variable computing costs that do not scale linearly. Even modest models using APIs can incur exorbitant cloud computing fees. Someone with business case development expertise can quantify the investments and returns. A worst case scenario is to be deep into a development project only to discover computing costs are far higher than anticipated.

4. ACCESS TO DEVELOPMENT TALENT

Speaking of development talent, finding quality AI/ML talent is even harder than finding Taylor Swift tickets. A strategist who can supplement your team with trusted AI/ML experts brings tremendous value. Domain-specific experience e.g. in marketing is a plus. This includes support in cloud infrastructure, data engineering, and security.

5. EXECUTIVE PRESENCE AND LEADERSHIP SKILLS

The risks here are very high because resistance can stall adoption if not properly addressed. Beyond domain expertise, the consultant must influence stakeholders to adopt AI. They should have a proven track record of crafting compelling yet non-threatening recommendations during discovery, development, and implementation. Alienating leadership teams during discovery is counterproductive. Credibility with executives and engineers is crucial. The wrong message will destroy a team’s morale and commitment but helping them develop new skills can boost careers. Talk to references, and ask lots of behavioral questions.

6. Training and change management acumen

AI brings profound changes. The strategist should have demonstrated experience in preparing teams to implement technology via training and change management. In marketing analytics, this entails upskilling analysts on new tools. They should explain complex topics clearly to anyone who asks, without alienating jargon. What change management frameworks do they use?

the tL; DR:

  • Don’t be blinded by influencer status: look for practical and deep domain expertise first.

  • Look for current knowledge of cloud computing costs, ethical issues in ai, and data security and privacy risks.

  • Insist on quality leadership and change management experience to ensure a smoother adoption curve.

  • Be practical and start small with a pilot that allows for reallocation and training of existing team members