Tips for Success as an AI Consultant

I burned thousands of dollars building 20+ AI automations. Here are the 6 brutal lessons that will save you weeks of wasted time: 1. Prioritize ruthlessly Don't automate everything that moves. I spent 3 weeks building an automation that saved 30 minutes per week 🤦♂️ Focus on bottlenecks that unlock your biggest goals first. 2. Document before you automate 90% of clients had vague process understanding. I tried reverse-engineering. It failed every time. Always map the process completely first. 3. Make success measurable "If you can't measure it, you can't improve it." Define quantitative metrics that are binary (good/bad). Vague success definitions kill AI projects. 4. AI isn't always the answer Sometimes the solution is simpler than you think. Focus on business value, not what's technically interesting. Ground clients in reality, not AI hype. 5. Keep it simple stupid I built a 5,000-line system with 83% accuracy. Ripped it out. Used one model. Got 97% accuracy in an hour. Complexity kills performance. 6. Set hard spending limits One coding error cost me $100 in 60 minutes. Set API caps at the provider level when possible. Build usage tracking into your code. The biggest mistake? Not learning from others' failures first. --- P.S. If you want my free 30-day AI insights series, comment "Purple Unicorn" below. 🦄

Stop treating ChatGPT like a search engine. It's a strategic consultant. "The quality of your questions is the ceiling of your output." One of the reasons I came to Shopify was to get extremely deep into AI. I've only been here 2 months now, but I've already learned more in the last 60 days about the topic than my entire life previously. This guy Chris Koerner is absolutely surgical with his business ideas and recently has been a go-to follow for me to use AI more efficiently, now that I have some good 101 understanding under my belt. He's geared towards SMB Entrepreneurs which is so up my alley, but I've applied his learnings big time in the Enterprise software world. For example, his latest 20 min mastermind (https://lnkd.in/eRNN4_27) is jammed pack with things I've immediately used this week like: 1. Stop asking for facts, start asking for strategy His example: Instead of "What are some good business ideas?" → "What are eight off the radar business ideas that people are talking about in message boards and in subreddits that are poised to explode over the next few years?" For me: "What are the top pain points manufacturing executives are discussing in industry forums that indicate they're open to evaluating new commerce platforms in the next 12 months?" 2. Feed it real context, not generic requests His example: Instead of "Write this email more simply" → "Write this email so a fifth grader could understand it." Instead of "Use good copywriting techniques" → First ask "What are some good copywriting techniques?" then pick the ones you want implemented. 3. Build repeatable workflows, not just prompts "Don't think about 'I need one good email' you want to think about 'I need a prompt that will write one good email anytime I need it to.'" 4. "What industries are notorious for having a bunch of one-star reviews where I could cold email owners and sell them a fix?" me: Perfect for sales prospecting - identify underserved markets in your vertical 5.: "Give me cool phrases from the Book of Mormon that don't show up anywhere else" (forcing ChatGPT to find unique, quality content) me: "Give me enterprise software implementation quotes that are absolutely gold based on what you know about digital transformation, but don't appear very often" - cuts through generic industry speak 6. "Here's what I have: a truck, time, and access to firewood. Give me a launch plan."  Perfect framework for resource optimization - "Here's what I have: $2M budget, 6-month timeline, team of 12 developers. Give me a market expansion plan." The switching costs of learning new AI workflows are massive, but the leverage once you get them dialed in? Game changing. (pictured below: the hilarious first-time output AI delivered me "show me a frustrated manufacturer trying to leverage AI" 🤪 )

𝐎𝐩𝐞𝐧𝐀𝐈 𝐞𝐧𝐭𝐞𝐫𝐢𝐧𝐠 𝐜𝐨𝐧𝐬𝐮𝐥𝐭𝐢𝐧𝐠 𝐢𝐬𝐧’𝐭 𝐛𝐨𝐥𝐝. 𝐈𝐭’𝐬 𝐥𝐨𝐠𝐢𝐜𝐚𝐥. APIs don’t drive transformation. Outcomes do. And enterprises don’t renew based on eval scores. They renew based on results. If you’re charging $10M or more, you’re not selling inference. You’re being hired to make change happen. → Outcomes aren’t model metrics. They’re hard-dollar results tied to business levers: • Revenue lifted • Costs reduced • Risk mitigated • Speed increased And they must be traceable, repeatable, and defensible to the CFO. No one cares that your model hit a benchmark if it didn’t move the business. → Governance, risk, and ethics aren’t afterthoughts. These are board-level concerns. • Where is the data coming from? • Is the output auditable? • Can we explain this to regulators, auditors, customers, and internal compliance? If you can’t answer these, the deal won’t close. Worse, the rollout might get blocked midway. → The business process landscape is a mess. No two departments operate the same way. No two regions share compliance posture. Legacy systems don’t talk. Ownership is unclear. Metrics are misaligned. And the moment AI touches PII, financials, or regulated flows, it gets political. This is where implementation fails unless you have experienced operators. → Maturity is uneven within and across organizations. You’ll find teams using AI in production, others stuck in PoC mode, and others buried in risk committees. Some industries like tech, fintech, and digital commerce move fast. Others such as insurance, healthcare, and government proceed with caution. You need industry-specific strategies and stakeholder-aware narratives. → Procurement is not your user. Deals go through security reviews, architecture boards, legal, and budget gating. You need structured onboarding playbooks, integration blueprints, and clear ownership models. Enterprise buyers don’t pay for great demos. They pay for delivery certainty. → Integration is where most AI projects stall. You’re not deploying into a clean stack. You’re embedding into 15 years of SAP customizations, Excel macros, and middleware no one owns anymore. And your outputs must trigger workflows in ServiceNow, Salesforce, Oracle, and other platforms. That will not succeed without proper change control, testing protocols, and infra-aware architects. → Org dynamics are non-trivial. Every AI deployment changes someone’s job. That means: • Resistance from middle managers • Fear from ops teams • Political roadblocks throughout rollout You’ll need stakeholder mapping, incentive alignment, and change stories crafted for each layer. → None of this is a reason to avoid consulting. But it’s a reminder that consulting isn’t about building powerful systems. It’s about making those systems actually matter inside real-world environments. If OpenAI gets this right, it won’t just lead in model development. It will own the enterprise transformation play.

“The technology problem (around AI implementation) has largely been solved” because we have available tools at reasonable costs that require attainable skillsets. “It’s a matter of will and vision more than tech at this point, so let’s point our tools to solving business problems.” Justin C. Here are four key pieces of advice from Croft to get started on your AI journey. full article: https://lnkd.in/dkeSwNu5 1. DON’T START WITH AI. It sounds counter-intuitive, but instead, center the business and customers first, and then ask how the technology can support strategy and operations. Croft’s key insight is that “the technology problem has largely been solved” because we have available tools at reasonable costs that require attainable skillsets. “It’s a matter of will and vision more than tech at this point, so let’s point our tools to solving business problems.” 2. CHOOSE YOUR USE CASE TO GET A WIN THAT MATTERS. After aligning on a business-first approach, it is time to narrow further and choose the right initial project. It’s important to choose something manageable enough to succeed and get a win under your belt in order to generate momentum and learning behind your AI program. At the same time, it needs to be something that’s going to move the needle and people will care about. Croft advises to focus on use cases with quantifiable outcomes that align with important metrics and have enough impact to improve to the organization. 3. BUILD YOUR TEAM. Here is the paradox about most AI projects: the people who build it are often not the people who use it. To be successful then requires that you bring into the project the end-users, so they understand what you’re trying to accomplish, how it works and why it matters to them. It is critical to get them on board early and engage them often. Remember the metrics from rule 2? This impacts your team as well. You should be able to answer the following questions: How well is AI solving your use case? How will success be measured? How are we defining efficiency and effectiveness? Be transparent not only about how the AI works, and how it gets to the results. Once people have buy-in to the use case and technology, they'll start changing their behaviors, which will drive your metrics. 4. THE ABILITY TO EXPLAIN IS MORE IMPORTANT THAN ACCURACY. It's worth it to give up some accuracy to have a more explainable model. This idea of a black box is over. You need to be able to stand in front of your CFO and answer the question, “How did the AI come to this conclusion?” And you want your team to critically evaluate the AI output rather than blindly trusting the answers, and people just will not trust or engage with it over the long term. Rather, they should understand the business goal you’re trying to achieve and have some understanding of how these models work. You don’t need everyone to understand how it works at a high level, but you do need everyone to understand how accuracy is measured and reported.

Topic: Free AI Consulting Tips I frequently offer free 30 minutes AI consulting to many. I decided to share some basic key tips (based on industry insights and academic research) to everyone here: • Identify Pain Points and Clarify Business Objectives: - Identify specific challenges suitable for AI solutions, aligning with strategic goals for significant impact - Thoroughly research to understand the root causes of these challenges. - Do NOT focus on building AI in search of in search of a problem. • Gather and Assess Data: - Check the availability and quality of your data. This is essential for training and deploying AI models effectively • Evaluate Organizational Impact and Focus on Business Value: - Assess where AI can deliver significant organizational impact, targeting areas ripe for automation or analytics improvements.  - Prioritize AI projects that clearly contribute to business goals (impact and value), such as enhancing patient/customer satisfaction, LOS, boosting revenue, cutting costs, or optimizing operations.  - Track the ROI of AI efforts to ensure alignment with overarching business objectives. • Demystify Technology: - Many overlook this critical point especially with regards to non-technical stakeholders.  - Simplify the explanation of AI's role and benefits and emphasize practical outcomes and solutions provided by AI. - Steer the conversation towards how AI solves specific business problems rather than building AI in search of in search of a problem. Disclaimer: Most of my AI work have been in healthcare #AIImplementation #AIOperationalization #AIInsights #AIforBusiness #Innovation #AIConsulting #DataDrivenDecisions

If you are selling SaaS, services, or anything for that matter, the days of selling via value propositions are over. Today, if you want to have consistent success, you must be highly competent at insight selling with a heavy focus on problem-solving capabilities. If you are not: 1) Teaching your potential customers something new about the industry. 2) Showing high competence in their day-to-day and unique problems that prevent them from succeeding. 3) Deeply understanding how your champions are goaled and measured. 4) Understanding their business, including operational workflows and competitors. 5) Showing them how you solve one or two critical problems that drive results they care about. 6) Showing them how they can be the hero by onboarding your solution. You will have a tough time succeeding in the new age of sales. So many sales reps simply "wing" their meetings, with zero prep, armed with a 35-slide deck about how great their company is and a "sales methodology." If this is you, you've already lost and will continue to get lapped by the TRULY consultative seller. Consultative selling isn't just a buzzword. It's a deliberate choice to be a learned practitioner in the industry your solution provides answers to problems for, and articulating that better than anyone could within your potential clients' business. With the abundance of AI tools now available to folks, there is no excuse other than inaction for not knowing points 1-6 for every potential client you speak with. Always come to the meeting with at least an attempt at these insights. Even if you are wrong, it will show that you are a student of your potential clients' world, and you will immediately propel yourself above the majority of others attempting to win their business. If you're looking for a jumping-off point, I've attached a high-level Evaluation Cycle Prep guide you can use to start thinking and acting in a truly consultative way. If you'd like further clarification on this guide I've written, please feel free to DM me or drop a question in the comments. Happy Selling, and make it a great day!

The organizations seeing the biggest wins with AI? They all started in the same place (and it’s not AI). After working with companies across industries, I've noticed a clear pattern in what separates successful AI initiatives from the rest. The companies that win with AI follow this path: 𝟭. 𝗧𝗵𝗲𝘆 𝗮𝗱𝗱𝗿𝗲𝘀𝘀 𝘁𝗵𝗲 𝗽𝗿𝗼𝗯𝗹𝗲𝗺 𝗳𝗶𝗿𝘀𝘁. What exactly are we trying to solve? Why haven't we been able to solve it before? What's the real business impact? 𝟮. 𝗧𝗵𝗲𝘆 𝗲𝘅𝗽𝗹𝗼𝗿𝗲 𝗽𝗿𝗼𝘃𝗲𝗻 𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻𝘀 𝗳𝗶𝗿𝘀𝘁. Before jumping into AI, they ask: Are there other tried-and-true technologies that could actually solve this problem? Could standard SQL scripts work? What about robotic process automation? 𝟯. 𝗧𝗵𝗲𝘆 𝗳𝗮𝗰𝗲 𝗔𝗜’𝘀 𝗿𝗲𝗮𝗹𝗶𝘁𝘆 𝗵𝗲𝗮𝗱-𝗼𝗻. Here's what I'm seeing in enterprise: AI agents still need very clear and precise rules to make decisions on their own. They're not the autonomous problem-solvers we sometimes imagine. 𝟰. 𝗧𝗵𝗲𝘆'𝗿𝗲 𝘀𝘁𝗿𝗮𝘁𝗲𝗴𝗶𝗰 𝗮𝗯𝗼𝘂𝘁 𝘁𝗶𝗺𝗶𝗻𝗴. If there are no other viable options after exploring existing tools and technologies that have already withstood the test of time, then, they explore AI. Don't use AI for the sake of AI. Start with the problem. Understand why traditional approaches haven't worked. Then choose the right tool for the job. This approach isn't about avoiding AI, it's about setting your AI initiatives up for real success. What's been your experience? How are you approaching AI strategy in your organization?

In a world where access to powerful AI is increasingly democratized, the differentiator won’t be who has AI, but who knows how to direct it. The ability to ask the right question, frame the contextual scenario, or steer the AI in a nuanced direction is a critical skill that’s strategic, creative, and ironically human. My engineering education taught me to optimize systems with known variables and predictable theorems. But working with AI requires a fundamentally different cognitive skill: optimizing for unknown possibilities. We're not just giving instructions anymore; we're co-creating with an intelligence that can unlock potential. What separates AI power users from everyone else is they've learned to think in questions they've never asked before. Most people use AI like a better search engine or a faster typist. They ask for what they already know they want. But the real leverage comes from using AI to challenge your assumptions, synthesize across domains you'd never connect, and surface insights that weren't on your original agenda. Consider the difference between these approaches: - "Write a marketing plan for our product" (optimization for known variables) - "I'm seeing unexpected churn in our enterprise segment. Act as a customer success strategist, behavioral economist, and product analyst. What are three non-obvious reasons this might be happening that our internal team would miss?" (optimization for unknown possibilities) The second approach doesn't just get you better output, it gets you output that can shift your entire strategic direction. AI needs inputs that are specific and not vague, provide context, guide output formats, and expand our thinking. This isn't just about prompt engineering, it’s about developing collaborative intelligence - the ability to use AI not as a tool, but as a thinking partner that expands your cognitive range. The companies and people who master this won't just have AI working for them. They'll have AI thinking with them in ways that make them fundamentally more capable than their competition. What are your pro-tips for effective AI prompts? #AppliedAI #CollaborativeIntelligence #FutureofWork

Most of the consulting I've been doing over the last few weeks has been related to setting up chatbots, generative AI for the front and back end of the service journey or knowledge management, and other automation tools for CX. When I come into these conversations, I first want to know what problem they're trying to solve and why they want to do it with automation. There are many problems automation can help you solve, but I’m finding too many people want to use it to replace a larger percentage of their workflow than is probably healthy for their customer experience. It CAN save time and money, but it still takes someone (or many people) to manage to make sure users are having the experience they deserve. Some Common Examples: Chatbots: A conversational chatbot requires constant management - your product and services CHANGE, so the chatbot needs training, new prompts, new decision trees, new conversation flows, etc. When you let them go stale they create infuriating, looping experiences for your customers. Auto-responders: Great when you want to let people know you received their request, trigger an update if wait times are longer than expected, or anytime you know your auto-response is 100% relevant to whatever action the customer took. When the path to contact support is a maze, users will take whichever path will get them to a text box - you can’t be certain they’re using the correct category, and then create an auto-response specifically for that category. Ticket Deflection: This usually comes in the form of serving knowledge base articles before a customer can reach out to support for a self-serviceable task. Again, this is great and can reduce your queue to mostly inbounds that require interaction from a person, but if you’re not keeping your KB content up to date, it’s useless and creates a headache fast. Be smart about the automation you're introducing to your service journey and make sure they're serving the customer and your team.

I love using Anthropic's Claude as a thought partner. Recently, "we" have been having weekly reflection sessions. I feed the 5 most interesting meeting transcripts from my past week into Claude's enormous context window and ask it to read them closely and ask me 3–5 hard-hitting follow-up questions. Then, I answer those questions with a long-form monologue of 45–90 minutes in length while driving or on a walk, pop the transcript back into Claude, and continue until we've gotten polished nuggets of insight. Here's what "we" learned this week after our latest reflection session, based on my work helping clients get their teams focused with OKRs and accelerate progress toward their goals with AI: 1. Automation enables augmentation. By delegating repetitive tasks to AI, you free up human attention/cognition to focus on higher-value efforts and innovation. 2. Integration with existing foundation models (GPT-4, etc.) is preferable to building custom AI initially in most cases. Leverage their scale while focusing your differentiation elsewhere. 3. Tight feedback loops and manual problem-solving early on is critical to ensure you are solving actual (vs. assumed) problems and properly designing solutions. No shortcuts. 4. Focus first on amplifying people’s strengths by delegating tasks they have to do but may not enjoy vs. attempting to outright replace roles. More humane and effective. 5. Specialized, proprietary data remains a key competitive advantage that should integrate with commoditized models. I highly recommend this exercise! It can help you make sense of lots of data in a short period of time, while remaining focused on the stuff that's most interesting, valuable, and deserving of your your most important resource—your attention.