Don’t Panic: What AI Can Really Do in Client Service
Management Summary
Part 2 of the series “Client Service of the Future” by Sahra Marquardt & Nina Huber
AI FOMO in Marketing: Why the Feeling of Being Too Slow Is Deceptive
I’m sitting in my old childhood bedroom in Berchtesgaden right now. When I look out the window, I see the mountains soaring majestically into the sky—motionless, completely unfazed by token limits, GPT updates, or automation workflows. Up here, the world feels wonderfully unhurried. And as I gaze at the peaks, I ask myself:
How many people here in the valley have actually ever really interacted with an AI?
Right after that, another thought crosses my mind—one that’s been on my mind even more:
How many people actually use AI much more efficiently than I do? And am I just way too slow?
Every day, we’re overwhelmed by a flood of new updates; LinkedIn is overflowing with “10 prompts that will change your life,” and I sometimes seriously wonder why I haven’t built my own app yet—one that organizes my entire daily routine at the push of a button. It feels as if AI development is racing past us in a high-speed train while we’re still standing on the platform trying to pull our ticket.
With this blog post, I’m trying to reassure myself—and maybe you, too—because we’re not alone in feeling this way.
A recent study by the Boston Consulting Group (see BCG: “AI Is Revolutionizing Marketing—But Most Companies Are Still Thinking Small,” 2025) actually paints a very reassuring picture of reality:
A full 85% of companies are still in the pilot phase or experimenting with individual tools. Only 15% have so far managed to integrate AI initiatives into their daily processes in a truly scalable way.
(See BCG: “AI Is Revolutionizing Marketing—But Most Companies Are Still Thinking Small,” 2025)
This means that while the “great AI revolution” is touted as a major development on paper, in practice—as we described in Part 1 —most marketing decision-makers are still just tentatively feeling their way up “Cringe Mountain.”
So, that’s where we’ll start. We don’t need to develop our own app to become more productive; instead, we need to cleverly integrate existing tools into our daily work routine. And here’s our very pragmatic roadmap for using AI as our co-pilot to take the pressure off us:
Account Management—From Data Search to Real-World Collaboration with AI as a Co-Pilot
We’ve reached a point where we’re slowly beginning to grasp the full scope of what’s technically possible, and it’s often this very realization that initially leaves us feeling overwhelmed. The areas where we can use AI to our advantage are virtually endless. And now it’s important to keep a cool head instead of panicking. We don’t have to reinvent the wheel or start building our own apps right away; instead, we need to first make sense of the many small opportunities while keeping the big picture in mind.
So, how can we make good use of AI right now?
From Searching to Asking Questions: Smart Prompting Workflows for Customer Interaction
It starts with the basics. We’ve stopped feeding AI—like search engines—with rigid search terms. Instead, we use it like a colleague: we ask questions, have email drafts refined, have phrasing varied, or have quick research summarized.
| In the past: Search engine | Today: AI prompt |
|---|---|
| “Benchmarking CPM for Display Marketing in E-commerce in Austria” | “I have a call tomorrow with an e-commerce client from Austria. What are realistic CPM ranges for display campaigns right now, and what three arguments can I use if the budget is called into question?” |
The difference is, we no longer have to search for individual links that we have to painstakingly click through. We get a ready-to-use outline for our client meeting right away.
A typical everyday scenario: Instead of spending 15 minutes polishing a diplomatic but firm rejection or deadline extension for clients, we type the key points into the AI and prompt it: “Write this in a polite, courteous, and solution-oriented way.” In seconds, those three bullet points turn into a finished draft that we just need to fine-tune.
Contract Analysis with AI: Quickly Identify Scopes and Deadlines
Anyone who works extensively with contracts knows the struggle: 30-page documents, densely printed, and you need an answer right now regarding a specific deadline. In the past, that meant pressing CTRL+F, scrolling through paragraphs, and hoping not to overlook anything. Today, it no longer takes an entire workday to extract the relevant information. We use AI as an intelligent reading assistant (of course, exclusively in a data-protection-compliant, secure environment!—but more on that in Part 3). We feed the document into the system and ask directly: “What is our exact scope?”, “What payment terms were agreed upon?”, and “Which deliverables does the team need to pay special attention to?”. Within minutes, this generates a clear onboarding sheet for the project team, and everyone knows immediately what needs to be done without wasting valuable hours poring over the contract.
Pitch Perfect: AI for Quick Competitive Analysis
Not only is everything around us moving faster, but we’re moving faster, too. Essentially, the most intensive part of a strategy lies in thorough research—if you understand the market, you’re already halfway there. It’s no secret that this often takes time we don’t actually have in our day-to-day pitch work. Traditional competitive analyses are essential for understanding market dynamics and the target audience’s pain points. That’s why we use AI to boost our research:
In the past, it used to take us hours to manually gather key points by browsing Google, annual reports, competitors’ social media channels, and industry forums, whereas today we can easily analyze the positioning of different providers using an AI prompt. We can draw on online presence and customer reviews to better assess the various competitors and evaluate where the greatest potential lies for positioning or an untapped niche.
While this doesn’t replace your own strategic thinking, it does provide, within minutes, a solid foundation for understanding market changes, competitor activities, and initial brainstorming on additional opportunities.
The “5th Eye”: AI as a Sparring Partner for Presentations
Working according to the dual-review principle is an absolute no-brainer in our line of work. Today, we also rely on AI as a sparring partner before our client meetings. We submit our briefing and our detailed slide draft and ask for feedback: “Have we really covered all of the client’s points?”, “Where are there gaps in our reasoning or potential for misunderstandings?”, or “What critical questions might decision-makers have about this presentation?” The AI doesn’t just proofread—it plays devil’s advocate—and makes our strategy deck even more watertight before the presentation.
AI in Project Management: Automated Transcription and Meeting Minutes
When I talk to our project managers, they’re passionate about roadmaps, creative scheduling, and orchestrating the team. What’s definitely not on their list of favorite tasks: typing up meeting minutes, manually creating tasks in project management tools, and searching for available time slots. It’s precisely this administrative burden that’s being taken off our hands today. Transcription tools like Gemini, Fathom, or Otter listen to calls and summarize what’s said.
Here’s a real-world example: Instead of spending 45 minutes sorting through notes after a 60-minute kick-off meeting, the AI immediately generates a structured summary that includes action items. Using prompts, these to-dos can then be converted into the appropriate format for Jira, Asana, etc., and imported.
But where are the limits? An AI understands spoken words, but not necessarily the subtext. It doesn’t recognize when a customer hesitates on a topic, or which tasks need to be strategically prioritized. Furthermore, there’s a clear guideline here: transcription tools may only be used with the explicit consent of all participants in the call, and sensitive customer data must remain protected.
The protocol is just a draft—approval and tact remain 100% human. This saves project managers time and finally gives them back the space to focus on what truly makes a project successful: managing the project and the people behind it.
I look out the window again. The mountains are still there, and the ICE has probably moved on a bit in the meantime—and that’s okay.
After all, how many people in the valley have actually worked with AI yet? Probably fewer than LinkedIn would have us believe. And of those who have, hardly anyone has built their own app. Most of them just got started: with a question, a contract, or a protocol. You don’t have to be faster than everyone else. You just have to take the first step.
A Look Ahead to Part 3: Where AI Reaches Its Limits in Client Service
AI can do a lot, but it can’t do everything. In Part 3 of our series, we’ll take a look at the far from insignificant flip side of the coin. Where does AI hit its hard limits when it comes to customer interactions? We’ll discuss failures, data protection safeguards, hallucinations—and why authenticity, empathy, and gut instinct will remain irreplaceable in consulting well into the future. Stay tuned!
