Guides
Guides written from real deployments: what works for B2B companies with 15 to 150 people, what it costs and where it usually breaks. If you do not know where to start, the first group is for you.
Book a free consultationGetting started
What can be automated, in what order, and how to prepare the company.
AI for a business owner: where do you start?
Your first 30 days with AI: how to pick one process in an hour, what your team does for free, where outside help starts and what to measure after a month.
What can be automated in your business?
Five traits of a process that is ready to automate, a five-question test, six areas of a company with examples, and four kinds of work we never touch.
Process automation step by step
How to map and automate one process: a week of watching instead of an interview, counted exceptions, five metrics read before the change, a parallel run.
Business process automation
What business process automation is, how it differs from RPA and AI agents, examples by department and how to pick the first process worth automating.
Workflow automation: the complete guide
The seven parts of a flow: trigger, conditions, actions, loop, human approval, error handling and a log. A failure table and the switch to production.
How do you prepare a company for an AI rollout?
AI readiness in three layers: data, people and rules. A checklist for each, a table of where data fails, and the point at which you hold off.
Digital transformation: where do you start?
Digital transformation without a three-year programme: five maturity stages, the symptom of each one, the first sensible move and a two-quarter plan.
AI agents and language models
What an AI agent is, how to implement one, and how to keep hallucinations out.
What is an AI agent?
An AI agent is a language model wired into your systems. The definition without jargon, the five parts, the limits of its autonomy, a worked example.
How do you implement an AI agent in a B2B company?
Which processes suit an AI agent, a readiness test in five questions, who has to decide on your side and what to demand from an implementation partner.
LLMs in business: what they are for
A language model predicts text. That is where hallucinations start. The three levels companies use one at, and the tasks you take off it first.
Prompt engineering for business: how do you write prompts?
The five parts of a production prompt, an output format the next step can read, 10 to 20 hard test cases and versioning that outlives the author.
AI hallucinations: how do you avoid them?
Why a model invents facts, how to ground an answer in company data, how to design and test a refusal, and what to do after a wrong answer.
How to build an AI knowledge base
What belongs in an AI agent's knowledge base, how to write a document so it can be cited, why you split it into chunks, and who owns versions and access.
MCP: Model Context Protocol
MCP is an open standard for connecting an AI agent to company systems. How it differs from a plain API integration, what it costs and where its limits are.
AI chatbot vs traditional chatbot: what is the difference?
Rule-based chatbot or one on a language model? Seven differences in a table, the hybrid setup, when a model is the wrong buy, and what to measure.
Sales, marketing and customer service
Pipeline, CRM, email and customer service with AI inside.
AI in B2B sales: how do you use it?
Which AI use cases in B2B sales work from the first week, and which ones cost you domain reputation, a credible forecast and your customers' data.
Sales process automation: how do you put it in order?
Stage criteria a rule can check, what fires when a stage changes, the fields a forecast needs, and the build order: routing and follow-up first.
Email marketing automation
Consent, a clean list and SPF, DKIM, DMARC before the first sequence. What to measure instead of opens, and six mistakes that burn a sending domain.
Customer service automation: how do you set up the process?
Customer service as a process: every channel in one queue, categories you can settle from the text, two clocks, a knowledge base, escalation and measurement.
How do you choose a CRM for a B2B company?
A CRM selection method that survives the demo: map the process, split must-haves from nice-to-haves, test on 100 of your own records, count the hidden costs.
How do you integrate CRM with AI?
CRM and AI integration in order: audit fields and duplicates, one process per pilot, narrow access, API versus Make and n8n, measurement at 30 and 90 days.
Finance, reporting and data
Invoicing, reporting, BigQuery and measuring the return on automation.
How do you automate invoicing?
When the invoice should be created, which fields to check before the run, and how not to issue it twice. Numbering, credit notes, currency, manual queue.
How do you automate reporting?
Before you automate a report, agree one definition per metric and one source for every number. Definition disputes, cut-off times and five steps.
How do you automate onboarding?
Write your own onboarding path first: seven steps, deadlines counted from the start date, one role per step and access ready before day one.
BigQuery for business: what it is and how to use it
What a data warehouse is, three signals that BigQuery is worth it, how the bill actually works, where the data sits, and when a spreadsheet is still enough.
Automation ROI: how do you calculate it?
The ROI formula for automation, a worked example that pays back in month two, the full cost table and five mistakes that break the calculation.
How do you measure automation ROI?
The baseline before the change, metrics you read straight from a system, adoption, the 30 and 90 day readings and what to do when the result comes out worse.
Tools and security
Choosing a platform, no-code, APIs for non-developers, GDPR and AI security.
How to choose an automation platform?
Choose an automation platform by process, not by brand: three processes on paper, five decisive questions, the per-operation billing trap and a real pilot.
Zapier vs Make vs n8n: which one should you choose?
Three platforms on the same axes: time to first flow, billing unit, where the data goes, connectors, upkeep. A direct recommendation for three cases.
No-code and low-code for B2B companies
Who should build in no-code, what a business-critical workflow needs before go-live, where version history is missing, and when to rewrite it in code.
API for beginners: what is it?
What an API is without the jargon, what „we have an API” really means, how keys, rate limits and webhooks work, and six questions for your vendor.
AI security in business: what should you watch out for?
Your team already uses AI chats. Which data must not leave the company, company accounts instead of personal ones, and what to do after a bad paste.
AI and GDPR: a guide for companies
What data goes into the model, how to narrow the scope, who acts as processor, where data is processed and when a DPIA is due. Settle it before the pilot.
First 30 minutes
are free
We'll show you how BOND or the Personal Agent works on sample data. No generalities - a concrete case from your industry.
👤 You write directly to Paweł
No reception, no helpdesk, no tickets. I'll answer directly: whether this makes sense for your company and what the first step would be.