AI consulting: where does AI actually help your business?
I look at how you work, test suitable AI tools against your real data and tell you what is worth it and what is not.
I start with your work, not with a tool. If a tool fits, I connect it and show your team how to work with it. If nothing fits, I build it.
30 minutes, free, booked through my calendar. You tell me where you are stuck with AI right now.
What is AI consulting?
AI consulting means someone looks with you at where your business loses time and checks whether AI can help. AI here means programs that read and write text, understand speech or create images.
With me, it does not end with a recommendation. I test the tools that could work against your real data, connect the one that fits and show your team how to work with it.
What you get at the end is not a slide deck, but a process that runs, or a reasoned answer that AI is not worth it here.
Where the first steps with AI stall
Too many tools, too little clarity
New ones appear every week. Which of them fits the way you work isn't in any comparison chart.
Accounts nobody uses
Signing up for a tool is quick. Without a set place in the daily routine, it goes unused.
Advice that ends with the slide deck
Recommendations are plentiful. Connected, set up, and explained — that part is missing.
From gut feeling to decision
- 1
Intro call
30 minutes, free. You tell me where things get stuck.
AfterwardsWe know whether it is worth going on.
- 2
Look at your processes
I map your processes, pain points and goals before we look at a single tool.
AfterwardsThe processes that cost you time are on a list.
- 3
Gather options
Which tools or models could work? And where does nothing off the shelf fit?
AfterwardsA few candidates instead of a long list of tools.
- 4
Test with your data
The candidates run with your real data and processes, not with made-up examples.
AfterwardsYou see what holds up and what does not.
- 5
Roll out and train
I connect the tool that fits and show your team how to work with it.
AfterwardsYour team can carry on without me.
In the intro call, we work out whether the next step is worth it for you.
AI strategy means getting the order right
An AI strategy does not have to be a thick report. With me, it is the answer to one question: which processes do you tackle in which order, and why?
The order follows from looking and testing: how often a process happens, how much time it costs and whether the test with your data holds up. Then you start with the first one, not with all of them at once.
A strategy that only exists on slides changes nothing about your work.
From writing tools to content production
Writing tools for marketing
Find the AI writing tool that suits your marketing team and set it up.
Models for customer service
Check which AI model answers customer enquiries well enough before you commit.
Document processing
Have documents read automatically instead of typing the data in by hand.
Images in creative workflows
Build AI image generation into your creative team's workflows.
Business figures
Build a custom dashboard that analyses your business figures with AI.
Office tasks
Let a workflow with AI handle recurring office and admin work.
Company knowledge via chat
A chat assistant that answers questions with your business's own knowledge.
For agencies and media companies
Set up AI-assisted content production for agencies and media companies.
Advice that doesn't stop at the slide deck
The right tool, used the right way
First your processes, pain points and goals, then the tool that fits. No off-the-shelf recommendations, nothing more complicated than necessary.
Hands-on implementation
I don't deliver a PowerPoint presentation. I test the tools against your real data, set them up and connect them to your programs.
Custom development when needed
If nothing on the market fits, I build it. Custom AI tools to your exact specifications — and the code belongs to you.
Your team can do it afterwards
I show your team how to work with the tools: from a good prompt, meaning the instruction you give the AI, to a better process.
Models stay replaceable
AI changes quickly. I build things so that a model or tool can be swapped later when the market changes.
Measure first, then switch
I check whether something is worth doing before anything else. When I switched image models, I measured speed, cost and result — and kept the cheaper batch route where it fits.
Attila Arndt
Triple A Digital, Cologne
When I advise against AI
AI is not always the answer. If it is not the answer for you, I will tell you.
The process is not clear yet
If a task runs differently every time, AI helps little. Then we sort out the process first, and AI comes after.
A simple program is enough
Sometimes a rule in your email program or a ready-made tool solves the problem. Then you need no AI, just the pointer.
The test does not hold up
If a tool does not work well enough with your real data, I tell you so instead of rolling it out anyway.
You only need slides
If the goal is a presentation for the drawer, I am the wrong person. My consulting aims at something that runs day to day.
Feel free to ask me in the intro call where AI would not help you, too.
Technologies used
OpenAI GPT · Anthropic Claude · Claude Code · OpenAI Codex · Midjourney · Flux · Stable Diffusion · Whisper · n8n · Make · Python · TypeScript · Next.js
Further reading
What you'll want to know first
How much does AI consulting cost?
The intro call is free and takes 30 minutes. What comes after depends on how many processes we look at and whether I build something. We settle that in the intro call.
Will I just get a recommendation at the end?
No, not a slide deck. I test suitable tools against your real data and processes, set them up, connect interfaces, and pass the knowledge on to your team.
What is the difference between AI consulting and an AI strategy?
The strategy is one result of the consulting: the order in which you tackle your processes, and the reason for it. The consulting goes further and also rolls out the first process.
Do I need to know about AI?
No. You know your processes, and that is enough. What you should bring is a contact person who knows the process, and access to the programs involved.
Do you need access to my data?
For the test, yes: I test with your real data and processes, not with made-up examples. Which data that is and how I get it, we discuss in the intro call.
Consulting first, or build straight away?
If the intro call already makes clear which process is worth it and what it needs, we can move straight on to a prototype. If not, we start by looking at your processes.
Roughly how much does it cost?
That depends on what needs to be built. We work it out in the free intro call.
How long does it take?
You get a first working prototype in days, not months.
What happens in the intro call?
Half an hour, free, booked through my calendar. You tell me which process is costing you time today; together we look at which one is worth tackling first — and whether I'm the right person for it.
Attila Arndt · Triple A Digital, Cologne
What's holding up your first steps with AI right now?
Pick a time that suits you. In the intro call, we'll work out which process is worth tackling first — and whether I'm the right person for it.
Or email me directly: hello@tripleadigital.de · I reply within 48 hours.