Lisa M. Walker · AI in Practice
Not a capabilities flex. A real look at where AI helped, where it needed serious supervision, and what it still cannot do on its own. The tools got faster. The judgment did not come with them. Turns out that part is still on me.
Selected Examples
Consulting Project · 2025-Present
The work
Led brand strategy and identity for Dr. Jill Simms' ministry: positioning, voice, visual identity, brand guidelines. Then built the website in Wix myself, copy, IA, and full implementation. No agency handoff.
What I delivered
A complete brand system: vision, mission, values, tone of voice, logo direction, color palette, guidelines doc. And a live site that actually reflects the brand, not just describes it.
Where AI came in
Used it throughout discovery and drafting: exploring positioning language, running through messaging alternatives, getting to first drafts faster. The brand voice document went through a lot of iterations before it felt right, and AI helped move through those rounds quickly. It also suggested some directions that were very confident and very wrong. That happens. On the build side, used it to troubleshoot Wix questions and think through content structure. Every judgment call about what the brand actually stood for was mine.
What I learned: AI moves the drafting timeline way up. But brand clarity still comes from understanding what the work actually is. That came from the client conversations, not the chat window.
Personal Brand Project · 2025
The challenge
A static resume was not telling the full story, or the right version of it for each role. The same background needed to read as GTM strategy, campaign management, marketing operations, or partner marketing depending on the hiring need. That is a positioning problem, not a formatting one.
What I built
Multiple role-aligned resume versions, each with a distinct positioning angle. A live interactive site covering experience, case studies, and capabilities. GA4 tracking to understand how people actually engaged with the content.
Where AI came in
Used AI to pressure-test the same background through different role lenses, identify which language matched job description terminology most closely, and accelerate bullet refinement across versions. It helped me see positioning tradeoffs faster: what to lead with, what to cut, and what to surface only in context. Fair warning: AI will absolutely tell you your draft is great if you let it. You have to ask it to be mean. The final calls around what was honest, strategic, and actually compelling were mine.
What I learned: AI is a genuinely useful sounding board for positioning. It is also a yes-machine if you stop pushing. Keep pushing.
Retrospective · Cricket Wireless / AT&T
The original work
Led a year-long brand and performance test across four submarkets using control-store methodology, measuring impact on gross adds, consideration, and brand perception to inform acquisition strategy. Involved designing submarket selection criteria, tracking three metrics simultaneously, and translating a year of data into a clear recommendation for leadership.
What made it hard
Multi-market, multi-metric studies generate a lot of signal and a lot of noise. The hardest part was not the analysis. It was knowing what the numbers actually meant for strategy, and making that legible to stakeholders who had competing reads on the same data.
How I'd run it differently today
Research and setup: AI would have accelerated submarket profiling and competitive context significantly, pulling demographic and retail patterns, helping frame the selection criteria faster.
In-flight analysis: With three metrics running across four markets, AI would have been useful as a synthesis layer, spotting patterns across data sets, flagging anomalies, structuring interim readouts so the team spent more time on interpretation and less time in spreadsheets.
The final recommendation: AI would have helped with structure and narrative: first drafts of the "so what" framework, alternative framings, exec summary language. But the strategic call behind the recommendation, which signals to trust, what the business implications actually were, still required people who knew the market and the organization.
The honest takeaway: AI would have made this project faster and the outputs cleaner. It would not have changed what the work required most: reading a year of mixed signals and deciding what they actually meant for the business. That part is not a prompt. That is judgment built over time, and it is still not something you can outsource to a chat window.
Workflow
AI is fastest when the problem is clear and the inputs are well defined. The further upstream you go, into ambiguity, strategy, and judgment: the more it needs a human in the loop who actually knows what good looks like. I am that human. Mostly.
Honest takes
Not best practices from a LinkedIn post. Just things that turned out to be true after using these tools on real work, including a few things I had to learn twice.
AI agrees too readily
If you hand it a mediocre draft and ask if it is good, it will find reasons it is good. Every time. You have to ask it to find what is wrong, push back on the brief, or generate a competing version. It is a thinking partner, not a critic. It will not tell you the emperor has no clothes. That is still your job.
Speed is real, but clarity is not free
AI gets you to a draft faster. It does not get you to a clear strategy faster. Garbage in, polished garbage out. The thinking still has to happen before you open the chat window, not inside it.
It is best mid-project, not at the start
The most useful moments are iteration and synthesis, not ideation from zero. Once you know what you are trying to say, AI helps you say it better and faster. Starting from scratch with no point of view just produces faster noise. Confident, well-formatted noise, but noise.
Judgment is not a phase, it is the whole thing
Every place AI shows up in my workflow, there is a human decision wrapped around it: which direction to pursue, which draft to kill, which signal to trust. The tool compresses time. The judgment does not compress. I have tried. It does not.
The bottom line
And that is genuinely good news for people who built the judgment the hard way. Which, if you are reading this, might be both of us.
I use AI to move faster and iterate more. But knowing when a draft is genuinely right versus just good enough is still a human call. AI exposes whether you had strategic judgment to begin with. If you did, it makes you faster. If you did not, it makes you faster at being wrong. The tool is not the variable. The person using it is.