# Olively

> Independent AI product consulting. Fanni Csincsák helps product teams work out
> which of their AI ideas are real, then builds them. Twelve years shipping
> product in fintech, SaaS and regulated enterprise.

Olively is a one-person consultancy operating remotely across the EU, UK and
worldwide. The differentiating position: most AI projects are theatre, and the
scarce skill is telling which ones are not, before budget is committed.

## Original data

These figures come from Fanni Csincsák's own engagements and are not aggregated
from other sources:

- 40 AI proposals assessed and classified inside a regulated European enterprise
- Roughly 1 in 10 filtered out as not genuinely AI — proposals a rules engine,
  an existing system or a query already handled
- Around 509 hours a month of identified saving across the quantified use cases
- 24,000 users in under six months on a fintech platform shipped end to end
- $1M+ in follow-on work from the Jurisage / AltaML engagement

## Services
- [AI use case assessments](https://www.olively.io/use-cases): one assessment per use case with a verdict on whether it is genuinely AI and worth building.
- [Is AI fraud detection worth building?](https://www.olively.io/use-cases/ai-fraud-detection): verdict — genuinely ai — the false positives are the product.
- [Is AI email triage worth building?](https://www.olively.io/use-cases/ai-email-triage): verdict — usually automation with a label.
- [Is an AI onboarding assistant worth building?](https://www.olively.io/use-cases/ai-onboarding-assistant): verdict — this is knowledge search — build that.
- [Is AI sales forecasting worth building?](https://www.olively.io/use-cases/ai-sales-forecasting): verdict — not a model problem — a data problem.
- [Is AI resume parsing worth building?](https://www.olively.io/use-cases/ai-resume-parsing): verdict — genuinely ai — and not the same as screening.
- [Is AI sentiment analysis worth building?](https://www.olively.io/use-cases/ai-sentiment-analysis): verdict — prohibited in workplaces — weak elsewhere.
- [Is AI image tagging worth building?](https://www.olively.io/use-cases/ai-image-tagging): verdict — genuinely ai — buy it.
- [Is AI data entry worth building?](https://www.olively.io/use-cases/ai-data-entry): verdict — usually — but ask why it exists.
- [Is AI content generation worth building?](https://www.olively.io/use-cases/ai-content-generation): verdict — genuinely ai — with a disclosure obligation.
- [Is AI contract review worth building?](https://www.olively.io/use-cases/ai-contract-review): verdict — genuinely ai — with a human gate.
- [Is AI demand forecasting worth building?](https://www.olively.io/use-cases/ai-demand-forecasting): verdict — not new ai — it's forecasting.
- [Is AI churn prediction worth building?](https://www.olively.io/use-cases/ai-churn-prediction): verdict — genuinely ai — but the model isn't the gap.
- [Is AI document classification worth building?](https://www.olively.io/use-cases/ai-document-classification): verdict — depends on the category count.
- [Is AI lead scoring worth building?](https://www.olively.io/use-cases/ai-lead-scoring): verdict — usually automation with a label.
- [Is AI expense approval worth building?](https://www.olively.io/use-cases/ai-expense-approval): verdict — not ai — your policy is already rules.
- [Is AI knowledge search worth building?](https://www.olively.io/use-cases/ai-knowledge-search): verdict — genuinely ai — usually the best first project.
- [Is AI code review worth building?](https://www.olively.io/use-cases/ai-code-review): verdict — genuinely ai — narrow it.
- [Is AI test generation worth building?](https://www.olively.io/use-cases/ai-test-generation): verdict — genuinely ai — with one serious caveat.
- [Is AI translation worth building?](https://www.olively.io/use-cases/ai-translation): verdict — genuinely ai — buy it.
- [Is AI ticket routing worth building?](https://www.olively.io/use-cases/ai-ticket-routing): verdict — usually automation with a label.
- [Is AI invoice matching worth building?](https://www.olively.io/use-cases/ai-invoice-matching): verdict — split — extraction yes, matching no.
- [Is AI CV screening worth building?](https://www.olively.io/use-cases/ai-cv-screening): verdict — genuinely ai, and high-risk.
- [Are AI meeting summaries worth building?](https://www.olively.io/use-cases/ai-meeting-summaries): verdict — genuinely ai — but buy it.
- [Is an AI customer support chatbot worth building?](https://www.olively.io/use-cases/ai-customer-support-chatbot): verdict — genuinely ai — scope decides it.
- [AI and EU AI Act glossary](https://www.olively.io/glossary): 33 terms from AI product work and the EU AI Act, each with a plain definition and the misreading that costs most.

- [AI Reality Check](https://www.olively.io/ai-reality-check): one-week fixed-scope
  diagnostic. Every idea on an AI roadmap scored against the same six questions,
  returning three lists — what is real, what is theatre, what is blocked.
  €1,500 startup / €3,500 enterprise.
- AI Build: €500–700 per day, or €4,000–8,000 fixed scope.
- Fractional AI Product Lead: €6,000–8,000 per month, about a day a week, embedded.

## Work

- [Acquiring 24,000 users for a fintech lifestyle app](https://www.olively.io/work/cence): a card-vault brief redirected into a fintech lifestyle platform before the build started; 24,000 users in under six months on both app stores.
  a discovery product. 24,000 users in under six months.
- [Lifting AI adoption without retraining the model](https://www.olively.io/work/jurisage): a legal tech AI classifier moved into the workflow researchers already used; more than $1M in follow-on work with no change to the model.
  moved into the workflow researchers already used. $1M+ follow-on work.
- [Scoring 40 AI use cases in a regulated enterprise](https://www.olively.io/work/ai-governance): forty AI proposals scored inside a regulated European enterprise, roughly one in ten of them not AI at all.
  inside a regulated European enterprise.

## Tools

- [AI readiness scorecard](https://www.olively.io/scorecard): fourteen questions scoring how ready a team is to build AI, across five areas — what the AI is for,
  data and knowing it works, the payoff, risk and sign-off, and who runs it after
  launch. Returns a score out of 100, a per-area breakdown and the weakest area
  named. Instant result, no email required. Distinct from an AI maturity model:
  it scores whether a specific piece of work is ready to start, not an
  organisation's capability ladder.

## Blog

Forty-one pieces. The recent ones are about AI decisions; the older set covers
startups, product decisions and design principles, written when Olively was a
product design practice.

- [How to tell an AI project from a project with AI written on it](https://www.olively.io/blogs/ai-project-or-project-with-ai-written-on-it): I scored forty AI proposals inside a regulated European enterprise. Roughly one in ten wasn't AI at all. The questions that caught them, before…
- [What the EU AI Act requires from 2 August 2026](https://www.olively.io/blogs/eu-ai-act-august-2026): Article 50 transparency obligations became enforceable on 2 August 2026. The Digital Omnibus delay applies only to high-risk systems.
- [Why most AI pilots never reach production](https://www.olively.io/blogs/why-ai-pilots-never-reach-production): what the 95% figure actually measures, why pilots stall, and the questions that separate the ones that ship.
- [What an AI readiness assessment actually covers](https://www.olively.io/blogs/what-an-ai-readiness-assessment-covers): What a credible AI readiness assessment examines, what it should produce, and how to tell a diagnostic from a sales process wearing one as a costume.
- [What to ask before hiring an AI consultant](https://www.olively.io/blogs/what-to-ask-before-hiring-an-ai-consultant): Seven questions that separate a consultant who ships from one who bills for slides — and what a weak answer to each one actually tells you.
- [How to scope an AI pilot so it can actually ship](https://www.olively.io/blogs/how-to-scope-an-ai-pilot): Most AI pilots fail on scope rather than technology. One measurable number, one workflow, and kill criteria agreed before the build starts.
- [What AI consulting costs, and what changes the number](https://www.olively.io/blogs/what-ai-consulting-costs): Published prices for AI consulting, what sits behind each number, and the four things that legitimately change a quote.
- [How to Build an AI MVP That Reaches Production](https://www.olively.io/blogs/how-to-build-an-ai-mvp): Most AI pilots never reach production. How to scope an AI MVP around one measurable outcome, decide if the problem needs AI at all, and set kill…
- [How CEOs Undermine Their Own User Research](https://www.olively.io/blogs/as-a-ceo-this-is-how-you-can-completely-mess-up-research-for-your-digital-startup-business): Three-quarters of startups fail, and culture is usually upstream of the reason. How founders distort research without meaning to, and what to do…
- [Building Digital Products in a Recession](https://www.olively.io/blogs/building-digital-products-in-a-recession): Downturns change what gets funded and what gets cut. How to keep shipping when budgets shrink, and which investments still pay back.
- [The Law of Similarity in Product Design](https://www.olively.io/blogs/the-law-of-similarity-in-product-design): elements sharing colour, shape or size are read as related — and what happens when genuinely different things end up looking alike.
- [The UX laws worth knowing, and when each one applies](https://www.olively.io/blogs/ux-laws-in-product-design): seventeen principles grouped by the decision each one helps with — grouping, cognitive load, interaction cost, expectation and memory.
- [Gestalt Psychology in Product Design](https://www.olively.io/blogs/gestalt-psychology-in-product-design): People read interfaces as groups before they read them as parts. The Gestalt principles behind that, and how to use them without over-designing.
- [Hick's Law: Why Too Many Options Slow Users Down](https://www.olively.io/blogs/hicks-law-in-product-design): Every extra choice adds decision time. What Hick's Law measures, where it applies in product design, and when reducing options costs you more…
- [How to Boost Sales for a Digital Startup](https://www.olively.io/blogs/how-to-boost-sales-for-your-digital-startup-business): Sales problems are often product problems wearing a disguise. Where digital startups lose revenue, and which fixes are worth doing first.
- [How to Build a Fintech MVP](https://www.olively.io/blogs/how-to-build-a-fintech-mvp): Fintech MVPs carry compliance, security and trust from day one. What belongs in the first version, what can wait, and what will block launch if…
- [How to Build a HealthTech MVP](https://www.olively.io/blogs/how-to-build-a-healthtech-mvp): Building in healthcare means regulation, clinical validation and privacy before you ship. What to scope into a HealthTech MVP and what to leave out.
- [How to Get Investors for a Digital Startup](https://www.olively.io/blogs/how-to-get-investors-for-your-digital-startup): What investors actually check before a first cheque, how to find the ones who fund your stage and sector, and the traction that changes the…
- [Startup Management Frameworks, Part 1](https://www.olively.io/blogs/how-to-increase-your-digital-startups-performance-by-using-the-right-management-framework-part-1): Frameworks fail when they are adopted whole rather than fitted to the team. Part one covers picking one that matches how your startup already works.
- [Startup Management Frameworks, Part 2](https://www.olively.io/blogs/how-to-increase-your-digital-startups-performance-by-using-the-right-management-framework-part-2): Part two covers putting a management framework into practice: what to measure, where teams stall in the first months, and how to tell if it is…
- [How to Motivate a Startup Product Team](https://www.olively.io/blogs/how-to-motivate-your-startups-product-team): Motivation in product teams comes from ownership and clear decisions, not perks. What actually moves a team, and what quietly drains it.
- [How to Pitch Your Startup and Get Funded](https://www.olively.io/blogs/how-to-pitch-your-startup-to-investors-to-get-funded): What belongs in a pitch deck, the order that holds an investor's attention, and the questions you should expect before the meeting ends.
- [How to Simplify a Digital Product](https://www.olively.io/blogs/how-to-simplify-digital-products-and-increase-value): Removing features is harder than adding them and usually worth more. How to find what to cut, and how to argue for it internally.
- [Jakob's Law: Users Expect Familiar Patterns](https://www.olively.io/blogs/jakobs-law-why-users-expect-your-product-to-work-like-the-others): People spend most of their time on other products and expect yours to behave the same way. When to follow convention, and when breaking it is…
- [Making Remote Work Work in a Downturn](https://www.olively.io/blogs/making-remote-work-work-for-your-digital-startup-during-recession): Remote teams cost less and break differently. What changes when budgets tighten, and which practices hold a distributed startup together.
- [Miller's Law and Working Memory in UX](https://www.olively.io/blogs/millers-law-enhancing-user-experience): People hold only a handful of items in working memory at once. What Miller's Law says, what it does not say, and how to chunk information…
- [Mission-Driven Business: Strategy and Design](https://www.olively.io/blogs/mission-driven-business-integrating-strategy-purpose-and-design-principles): A mission is only useful if it changes decisions. How to connect purpose to product strategy so it survives contact with a roadmap.
- [Occam's Razor in Product Design](https://www.olively.io/blogs/occams-razor-in-product-design): The simplest design that solves the problem usually wins. How to apply Occam's Razor to features and flows without stripping out what users need.
- [Parkinson's Law in Product Design](https://www.olively.io/blogs/parkinsons-law-in-product-design): Work expands to fill the time available, and so do interfaces. How Parkinson's Law shows up in scope, deadlines and product decisions.
- [The Peak-End Rule in Product Design](https://www.olively.io/blogs/peak-end-rule-in-product-design-and-business): People judge an experience by its most intense moment and its ending, not its average. What that means for onboarding, errors and cancellation flows.
- [Postel's Law in Product Design](https://www.olively.io/blogs/postels-law-in-product-design): Be liberal in what you accept, strict in what you send. How the robustness principle applies to forms, inputs and error handling.
- [The Law of Common Region in Product Design](https://www.olively.io/blogs/power-of-common-region-in-user-centric-product-design): A border or background makes separate elements read as one group. How common region works, and where it beats spacing alone.
- [Product Designer vs UX Designer: Which You Need](https://www.olively.io/blogs/product-designer-vs-user-experience-designer-which-one-do-you-need): The titles overlap and the roles do not. What each one actually does day to day, and which one your product needs at its current stage.
- [Product Manager vs Project Manager](https://www.olively.io/blogs/product-vs-project-manager-whats-the-difference): One owns what gets built and why, the other owns how it lands on time. Where the roles differ, where they overlap, and when you need both.
- [How Digital Startups Grow in a Recession](https://www.olively.io/blogs/strategies-for-digital-startups-to-grow-in-recession): Downturns reset who your competition is. Where growth is still available to digital startups, and which strategies stop working when spending…
- [Dieter Rams' Ten Principles, Applied to Product](https://www.olively.io/blogs/ten-principles-of-good-design-business-and-design-aspects): Rams wrote ten principles for industrial design. Read against digital products and business decisions, most of them still hold. A few need arguing…
- [Tesler's Law: Complexity Has to Live Somewhere](https://www.olively.io/blogs/teslers-law-balancing-complexity-in-product-design): Every system has irreducible complexity. The only question is whether the product absorbs it or the user does. How to decide where it belongs.
- [The Business KPIs That Matter in Digital Products](https://www.olively.io/blogs/the-business-kpis-that-matter-in-digital-products): Most product dashboards measure activity rather than value. Which KPIs connect to the business, how to set a baseline, and what to stop tracking.
- [The Doherty Threshold in UX Design](https://www.olively.io/blogs/the-doherty-threshold-in-ux-design): Below about 400 milliseconds, people stay engaged; above it, attention leaks. What the Doherty Threshold means for interface and perceived speed.
- [The Law of Proximity in Product Design](https://www.olively.io/blogs/the-law-of-proximity-in-product-design-enhancing-user-experience): Things placed close together are read as related, whether or not they are. How proximity groups meaning, and how spacing errors mislead users.
- [The Pareto Principle in Product Design](https://www.olively.io/blogs/the-pareto-principle-in-product-design): A small share of features carries most of the value. How to find which ones, and what to do about the long tail that remains.
- [Fitts' Law in Product Design](https://www.olively.io/blogs/the-power-of-fitts-law-in-product-design): Target size and distance decide how fast something can be clicked or tapped. What Fitts' Law predicts, and where it matters most on mobile.
- [The Law of Prägnanz in Product Design](https://www.olively.io/blogs/the-power-of-the-law-of-pragnanz-in-product-design): People resolve complex images into the simplest form available. What Prägnanz means for layout, icons and visual hierarchy.
- [Uniform Connectedness in Product Design](https://www.olively.io/blogs/the-power-of-uniform-connectedness-in-product-design): Elements joined by a line, border or shared background read as one unit. The strongest of the grouping principles, and how to use it.
- [What Does a UX or Product Designer Do?](https://www.olively.io/blogs/what-does-a-ux-or-product-designer-do): Beyond screens: research, problem framing, trade-offs and the decisions that shape what gets built. What the role covers in practice.
- [What a Product Design Consultancy Actually Does](https://www.olively.io/blogs/what-is-a-user-experience-product-designer-consultancy): What you get from a design consultancy, how engagements are usually scoped, and how to judge whether one is worth hiring for your problem.
- [What Is User Experience (UX)?](https://www.olively.io/blogs/what-is-user-experience): UX covers everything that happens between a person and a product, not just the interface. A plain definition, and why the distinction matters.
- [When to Hire a Product Design Consultant](https://www.olively.io/blogs/why-do-you-need-a-product-design-consultant-to-achieve-business-success): An outside consultant buys distance and speed, and costs continuity. When that trade is worth making, and when hiring in-house is the better call.
- [Why You Need a UX or Product Designer](https://www.olively.io/blogs/why-do-you-need-a-ux-or-product-designer): Design decisions get made whether or not a designer makes them. What changes when someone owns them deliberately, and what it costs when nobody does.

- [Integrating an AI platform into enterprise ITSM](https://www.olively.io/work/ai-platform-integration): three nested redirects on a vendor AI platform going into live enterprise service management: severity against urgency, phased rollout, multi-route integration.
- [Cutting enterprise change requests 15 to 7 minutes](https://www.olively.io/work/change-chatbot): an enterprise chatbot scoped to one path; change creation from fifteen minutes to seven and policy compliance from 72% to 89%.
- [Turning a per-client survey into one framework](https://www.olively.io/work/survey-framework): a per-tenant survey build redirected into a reusable multi-tenant product deployed by configuration.

## Archive

Eighteen case studies from before Olively became an AI consultancy. Kept because
they still rank and because the reasoning holds up.

- [SaaS Inventory App for Tracking Stock & Sales](https://www.olively.io/portfolio/saas-inventory-management-app): Sales rose 4% after launch.
- [Fintech Figma Design System](https://www.olively.io/portfolio/fintech-design-system): 1,728 components and 915 icons, documented in Zeroheight with code snippets alongside.
- [SaaS CRM Mobile App](https://www.olively.io/portfolio/saas-mobile-app): Invoices paid rose 23% and quote conversion 40%.
- [CSV Parsing Bank Statements](https://www.olively.io/portfolio/csv-parsing-bank-statements): Around 2,000 CSV formats were handled by launch.
- [Fintech Visual Identity & Branding](https://www.olively.io/portfolio/fintech-visual-design-and-branding): Building a brand from nothing for Canua, as the startup's first full-time hire, working directly with the CEO.
- [SaaS CRM Webapp (Dashboards)](https://www.olively.io/portfolio/saas-crm-dashboards): Completed onboarding rose 400% after the first concept was replaced, and invoices paid 23%.
- [Designing a Product Hunt Launch](https://www.olively.io/portfolio/designing-a-product-hunt-launch-fintech-startup): A Product Hunt launch for Canua, built on analysis of what had actually ranked rather than on taste.
- [Fintech landing page optimisation](https://www.olively.io/portfolio/fintech-landing-page-optimization): Onboarding conversion rose 36%, measured with an average of 460 participants per test.
- [Referral and review management web application](https://www.olively.io/portfolio/referral-and-review-management-web-application): Referrals rose 36% and leads 16%.
- [Streaming tablet app for kids](https://www.olively.io/portfolio/streaming-tablet-app-for-kids): An MVP delivered in ten days with a prototype the client could run usability tests against.
- [Patient-centric healthcare web experience](https://www.olively.io/portfolio/patient-centric-healthcare-web-experience): An 18% increase in activation on the booking process, measured against the old site with an average of 500 participants per test.
- [Tax automation for American expats](https://www.olively.io/portfolio/tax-automation): Sales rose 120% and conversion 360%.
- [E-commerce and healthcare: conversion optimisation and redesign](https://www.olively.io/portfolio/e-commerce-healthcare-conversion-optimization-and-redesign): Removing the intermediate cart step lifted overall conversion 78%.
- [Making remote connections meaningful](https://www.olively.io/portfolio/online-meeting-tool-for-teams-with-gaming-integration): Twelve features and four games, built on the finding that the problem is not a shortage of activities — it is that the….
- [Copy.ai Workflows: a homepage for a Gen AI platform](https://www.olively.io/portfolio/copy-ai-ai-powered-copywriting): Over 100,000 sign-ups, $2.
- [Column Tax: embedded tax experiences](https://www.olively.io/portfolio/embedded-tax-experiences): A tax return that can be set up and filed in under five minutes, and a brand simple enough to sit inside a partner's app….
- [Landing page for a machine learning startup](https://www.olively.io/portfolio/landing-page-for-machine-learning-startup): An end-to-end design vision the team could build from, with the model pipeline turned into something a visitor can watch….
- [Sola: bridging the gap, one disaster at a time](https://www.olively.io/portfolio/sola-insurance): Traffic rose 95%, page visits 53%, and average visit duration doubled.

## Legal

- [Privacy](https://www.olively.io/privacy): what is collected, why, how long it
  is kept. Analytics and session recording run only with consent.
- [Cookies](https://www.olively.io/cookies): nothing is set until you agree.
- [Terms of use](https://www.olively.io/terms)
- [Imprint](https://www.olively.io/imprint): Olively Kft., registered in Hungary.

## Contact

- [Contact](https://www.olively.io/contact): send a note, book twenty minutes, or email
  directly. Reply within two working days. Contracting is with Olively Kft., a Hungarian
  limited liability company, EU VAT HU32055164, invoiced in euros with reverse-charge VAT
  for EU business customers outside Hungary. NDAs signed as standard. No software is
  installed on client systems and no production data access is required.

hello@olively.io — https://www.olively.io