Question 1

How is yesfor.ai different from other AI consulting firms?

Accepted Answer

yesfor.ai does not deploy AI right away. First we diagnose the company's readiness, then we put processes and data in order, and only then do we deploy AI. We work in two models: „we build for you” (custom platforms based on your data, for companies without their own IT) and „we build with you” (we strengthen your IT team through mentoring, code review and mapping). MIT NANDA research (2025) shows that 95% of GenAI deployments deliver no ROI due to a lack of organisational preparation.

Question 2

Can I do the process mapping with my own IT team?

Accepted Answer

Yes, and we often recommend that model. Our „Internal IT support” service (2,500 - 7,000 zł / month) covers mentoring, code review and architecture advisory. You keep control and knowledge in house, we provide the lead. This approach works well for companies that already have strong IT and do not want to outsource core competencies.

Question 3

How much does the AI Readiness Audit cost?

Accepted Answer

The AI Readiness Audit costs from 8,500 to 24,000 zł net, depending on the size of the company and the scope. It takes 1-2 weeks. It covers a diagnosis of processes, data, team and ROI metrics. The result: an AI Readiness Score report (0-100), a list of priority blockers and a roadmap. For smaller companies we also offer a Mini audit (3-5 days, 4,000 - 8,500 zł). Polish public funding (KFS) can cover up to 80%.

Question 4

What is Shadow AI and how does it affect us?

Accepted Answer

Shadow AI is when employees use AI tools (ChatGPT, Claude, Copilot) without their company's knowledge or policy. Research shows that most employees do this regularly, pasting in client data, contract fragments and source code. That is a serious risk of data leaks and non-compliance with GDPR and the AI Act. Our Cybersecurity & AI Policy service (4,500 - 17,500 zł) covers a Shadow AI audit, design of the company policy and security training.

Question 5

Do you offer funding for training and deployments?

Accepted Answer

Yes. In Poland, most of our services qualify for public funding. KFS (the National Training Fund): up to 80% of costs for employers (standard), up to 100% for micro-enterprises (up to 10 employees), capped at 300% of the average annual salary per participant. PARP: up to 70% for SMEs, and training can be combined with deployment. ESF+: depending on the region. We help with the application paperwork. Example: team training priced at 5,000 zł net can cost your company 1,000 zł net. These programmes apply to the Polish market.

Question 6

Are AI agents worth the investment?

Accepted Answer

With caution. Gartner (June 2025) predicts that 40% of agentic AI projects will be cancelled by 2027, driven by hype, „agent washing” (vendors calling chatbots agents), poor decisions and a lack of deployment strategy. Of the thousands of vendors claiming agentic capabilities, realistically only around 130 have genuine agentic functions. Our position: AI agents can be valuable, BUT only in companies that have first put their processes and data in order. Without that, an agent escalates the chaos faster.

Question 7

Do you build platforms based on our data?

Accepted Answer

Yes. It is one of our core specialisations: custom platforms, applications and AI solutions built on a specific company's data. We have experience building multi-tenant SaaS platforms across different industries. Pricing is individual after a short conversation and an initial audit, because every platform means a different scope, different data and different integrations.

Question 8

Who is yesfor.ai for?

Accepted Answer

We work with medium and large Polish enterprises across different industries. The smallest projects start at 4,000 zł net (Mini audit, with KFS funding realistically below 1,000 zł).

Question 9

Why „order first, AI second”?

Accepted Answer

Research by MIT NANDA (2025) and Gartner (2026) shows that only 5% of generative AI deployments produce a measurable ROI. RAND Corporation (2025) identified three main causes of failure: data quality problems, organisational immaturity and scope drift. All of these are organisational problems, not technological ones. yesfor.ai starts by solving them.

Question 10

Do you work in a „we build for you” or „we build with you” model?

Accepted Answer

Both. For companies without their own IT (or with minimal IT), „we build for you”: full deployment, custom platforms, applications and AI solutions. For companies with strong internal IT, „we build with you”: mentoring, code review, architecture advisory, process mapping. We choose the model during a free intro call.

Question 11

Can I get KFS or PARP funding?

Accepted Answer

Yes. In Poland, most of our services qualify for public funding. KFS (the National Training Fund): up to 80% of costs for employers (standard), up to 100% for micro-enterprises (up to 10 employees), capped at 300% of the average annual salary per participant. PARP: up to 70% for SMEs, and training can be combined with deployment. ESF+: depending on the region. We help with the application paperwork. Example: team training priced at 5,000 zł net can cost your company 1,000 zł net. These programmes apply to the Polish market.

Question 12

Are the prices negotiable?

Accepted Answer

The ranges (for example 8,500 to 24,000 zł for an audit) are a starting point. A larger scope means a higher price, a smaller one a lower price. We negotiate scope, not an hourly rate. For custom projects (platforms, applications) the quote is individual, after a free call and a mini audit.

Question 13

Do you issue VAT invoices?

Accepted Answer

Yes. yesfor.ai operates as a Polish company, issuing VAT invoices with 14-30 day payment terms. Larger projects can be settled per milestone.

Question 14

How quickly do you respond to an enquiry?

Accepted Answer

During business hours (Mon-Fri, 9-17 CEST) within 4 hours. Outside those hours, within 24 hours. The form on /en/contact goes straight to Arleta's inbox.

Question 15

What happens after I book a call?

Accepted Answer

A 30-60 minute video call (Google Meet) or phone call, free of charge. No presentations, no slides. We ask about the problem, explain the method and check whether working together makes sense. After the call: a proposal within 48 hours, or an honest „we are not the right fit”.

Question 16

Is the 60-minute intro consultation paid?

Accepted Answer

No. It is free and with no obligation. It helps establish which model of cooperation makes sense (we build for you, we build with you, DIY with support) and which service is worth choosing.

Question 17

What is Shadow AI and how does it affect us?

Accepted Answer

Shadow AI is when employees use AI tools (ChatGPT, Claude, Copilot) without the company's knowledge. Most employees do this regularly. The consequences: leaks of client data, contract fragments and source code. A GDPR and AI Act risk. The fix: Cybersecurity & AI Policy (4,500 - 17,500 zł, 1-2 weeks), a Shadow AI audit, a policy and team training.

Question 18

Can I do the process mapping with my own IT team?

Accepted Answer

Yes. Our „Internal IT support” service (2,500 - 7,000 zł / month) covers mentoring, code review and architecture advisory. You keep control and knowledge, we provide the lead. For companies with strong IT.

Question 19

Are AI agents worth the investment in 2026?

Accepted Answer

With caution. Gartner (June 2025) predicts that 40% of agentic AI projects will be cancelled by 2027. The reasons: hype, „agent washing” (vendors calling chatbots agents), a lack of strategy, FOMO. Of the thousands of vendors claiming agentic capabilities, realistically only around 130 have genuine agentic functions. Our position: agents can be valuable, but only in companies that have first put their house in order.

Question 20

Do you build platforms based on our data?

Accepted Answer

Yes. It is one of our specialisations: custom platforms, applications and AI solutions built on your company's data. We have experience building multi-tenant SaaS platforms across different industries. The quote is individual after a short call and an initial audit.

Question 21

Which AI models do you use?

Accepted Answer

Vendor-agnostic. OpenAI (GPT), Anthropic (Claude), Google (Gemini), local models (Mistral, Llama). For each use case we choose the best model on cost, latency, quality and compliance. Often a mix of several models in a single application.

Question 22

Do you build on-prem solutions?

Accepted Answer

Yes, for companies with compliance requirements (finance, healthcare, legal). Local LLMs (Llama, Mistral) on the client's infrastructure. Costs are higher, but the data never leaves the company.

Questions about yesfor.ai.

Your questions matter too.