Natalie Torres spent three months using AI for goal advice before she got anything useful out of it.
Not for lack of trying. She’d read the articles, watched the YouTube tutorials, and had a paid subscription to two different AI tools. She was asking questions. The AI was answering. But the answers all felt like they belonged in a self-help book — reasonable, vaguely applicable, and completely unmoored from her actual life.
Natalie was a 41-year-old VP of Operations at a scaling Series B startup. She had twelve direct reports, two kids under eight, and a transition to think through: the company was approaching a size where her generalist role was likely to narrow, and she needed to either grow into a C-suite position or position herself for a lateral move to a different company. She also hadn’t exercised regularly in two years and had a writing habit she’d been meaning to restart since her last child was born.
Her AI goal advice looked like this.
Her prompt (Month 1): “Help me figure out my professional goals for the next year. I’m a VP of Operations at a startup and I want to grow in my career.”
AI response: “Great question! Here are some professional goals to consider as a VP of Operations:
- Develop your executive presence and leadership skills
- Build deeper expertise in a specific operational domain
- Grow your professional network
- Seek out a mentor or executive coach
- Identify high-visibility projects that demonstrate your strategic thinking…”
She recognized the irony later. She’d spent years telling junior employees that vague goals produced vague results. She was doing exactly that with her AI conversations.
“I kept thinking the AI just wasn’t good enough for this kind of thing,” she said. “I thought it needed a human to really understand a complicated situation. It took me a while to realize the AI wasn’t the problem.”
The Turning Point
The shift came when Natalie read a thread on LinkedIn about AI context documents — the idea that the quality of AI output is almost entirely determined by the quality and depth of context you provide.
She spent a Saturday afternoon writing hers. It took about 45 minutes, including the time she spent staring at the questions and thinking through honest answers.
Here’s a condensed version of what she wrote:
Identity: Analytical, systems-oriented, strong executor but sometimes slow to make decisions when I have incomplete information. I work best in the mornings — I’m genuinely useless after 4pm for anything complex. I tend to over-research before acting and can delay on decisions that make me uncomfortable. I have strong follow-through when I’ve committed publicly but drift on private goals.
Situation: VP of Operations at a 200-person Series B startup. We’ve grown from 80 to 200 people in 18 months. I have 12 direct reports. The CEO and I have a strong relationship but I’m aware the company is getting to a size where my generalist role will likely specialize. I have two kids (8 and 6). My partner works full-time as well. We have good childcare coverage during the week. My mornings before 7am are usually free. Evenings are family time — non-negotiable.
History: In my career, I’ve grown the most through stretch roles that forced me to build new skills under pressure. I’ve never successfully built a habit through gradual accumulation — I always need a clear trigger or forcing function. I’ve abandoned three different fitness programs because they required gym time I couldn’t reliably protect. I’ve started and abandoned journaling/writing four times; each time I stopped when life got busy.
Values: Work that requires me to think hard about complex problems. Being present with my kids while they’re young — I will trade career pace for that. Financial security matters to me, but not at the cost of everything else. I want to build something I’m proud of, not just accumulate a title.
Constraints: About 45 minutes before my day starts (5:45-6:30am). Very limited evening time. My company offers $2,000/year for professional development. I don’t have a strong network outside operations — this is a gap I’ve been aware of for years.
With this document in hand, she went back to AI with the same underlying question — professional direction for the year ahead. The output was unrecognizable.
Before and After: The Same Question, Two Different Conversations
With context:
“Based on everything I’ve shared, I need to think through my professional direction for the next 12 months. My company is growing in a way that may force my role to specialize, and I’m not sure whether to lean into the operational side or try to expand toward a COO path. What do you think the right goal is for me — and what concerns do you have?”
AI response (condensed):
“A few things stand out from your situation that I’d think through carefully before setting a direction.
The first is timing. At 200 people heading to 400+, you’re right that your generalist role will likely narrow — but which way it goes often depends less on your choice and more on the organizational structure the CEO decides on in the next 12-18 months. Before you invest heavily in either path, I’d suggest getting explicit about what the CEO is thinking. That conversation seems overdue given your close relationship.
The second is your stated constraint around network. A COO path at a different company requires a strong network that can vouch for your work. You’ve named this gap explicitly. Building that network is long-lead work — probably 18 months minimum before it pays off. If COO is a real option you want to keep open, starting the network work now (even slowly) is more important than any skill development.
Given your morning availability and your history of needing forcing functions rather than gradual habits, I’d suggest a 90-day goal that does two things: First, have the explicit career conversation with your CEO — you need this information before making any major bets. Second, commit to one external relationship-building action per week: a coffee, a LinkedIn engagement, a referral. Not because it will build your network fast, but because it will establish the habit while the stakes are still low.
What concerns me: your history of research before action is a real risk here. This is a situation where gathering more information will feel productive but delay the actions (conversations, relationship-building) that actually matter. I’d watch for that pattern specifically.”
Natalie read the response twice. “It addressed the thing I’d been avoiding,” she said. “I’d been thinking about what skills to build instead of having the conversation I needed to have with my CEO. The AI called that out directly because I’d told it I delay on uncomfortable decisions.”
The advice wasn’t magic. It was pattern-matching applied to her specific context. But it felt like talking to someone who actually knew her.
What Changed in Her Practice
Over the following three months, Natalie refined how she used AI for goal coaching. A few specific practices made the biggest difference.
She started every session by pasting her context document. Not because the AI needed to be reminded of every detail every time, but because it set the frame for the conversation — signaling that she wanted situationally specific advice, not general frameworks.
She used Beyond Time as her primary goal tool precisely because it maintained her context automatically across sessions. She didn’t have to paste anything. The accumulated context from previous conversations was always present. “The quality of advice after three months was noticeably better than the quality after three weeks,” she said. “Not because the AI got smarter — but because it knew me better.”
She explicitly invited criticism before asking for suggestions. Her standard opening evolved from “help me with [X]” to “before you suggest anything, what concerns do you have about [X] given what you know about me?” The reframing produced dramatically more useful responses.
She updated her context document quarterly. The version she wrote in January looked different by April: she’d had the CEO conversation (which revealed he was thinking about a COO role in 18 months), she’d built a modest network in operations leadership, and she’d established a consistent morning writing habit by connecting it to her existing morning coffee routine rather than treating it as a separate habit.
The Fitness Thread
Alongside the professional goals, Natalie also used the same approach to finally establish a consistent fitness habit — something she’d tried and failed at three times.
Her history layer was explicit: “I’ve abandoned three different fitness programs because they required gym time I couldn’t reliably protect. The programs I’ve tried assumed I had 45-60 minutes available on a reliable schedule, which I don’t.”
The AI’s suggestion: stop trying to rebuild a gym habit. Given her 45-minute morning window and history of schedule unpredictability, it recommended a home-based 15-minute mobility and strength practice — something sustainable below the threshold where life events typically derailed her. “You don’t need to be ambitious here,” the AI told her. “You need to be consistent at something modest. Every previous attempt has been derailed by ambition, not laziness.”
She’s been consistent for eight months. It’s the longest stretch of regular exercise since before her first child was born.
What This Illustrates
Natalie’s story isn’t exceptional — it’s representative of a pattern that appears consistently among people who go from generic to genuinely personalized AI goal advice.
The shift isn’t about finding a better AI tool or writing more sophisticated prompts. It’s about doing the work of self-documentation that makes personalization possible — and then treating AI as a thinking partner rather than a search engine.
“I thought the AI didn’t understand my situation,” she said. “The truth was I hadn’t told it my situation. Once I did, it was honestly more useful than most of the coaching I’ve paid for.”
The pattern holds: better context produces better advice. The AI was always capable. The limiting factor was the input.
For the framework Natalie used, see the Complete Guide to AI-Personalized Goal Advice. For the step-by-step process of building your own context document, see How to Get Truly Personalized Goal Advice from AI.
Frequently Asked Questions
-
Is this case study based on a real person?
Natalie Torres is a composite character based on common patterns observed in how professionals use AI for goal coaching. The situations, prompts, and AI responses are illustrative examples designed to show the practical difference between generic and personalized AI goal advice.
-
How long did it take Natalie to see a real difference in the quality of AI advice?
The shift was noticeable within the first conversation after building her context document. The deeper benefit — advice that drew on her accumulated history and updated situation — took about three months of consistent use to fully develop. Most people report a meaningful improvement from the first contextualized session.
-
Can someone without a technical background do what Natalie did?
Yes. The approach requires no technical skills — only honest self-reflection and willingness to have a substantive conversation. Natalie's initial attempts were generic because she was using AI like a search engine, not because of any technical limitation. The fix was entirely in how she approached the conversation.