Portals and information systems built on real data
Directories, price comparisons and internal systems where the content comes from a database instead of being typed page by page — so 600 pages cost about what 10 would.
What is a data-driven portal?
A portal is a website whose content lives in a database rather than in pages someone typed. You describe one page template, point it at a few hundred (or few hundred thousand) records, and every record gets its own proper page — with its own title, its own structured data and its own place in the sitemap.
That changes the economics. A hand-built page costs the same as the one before it, so a 600-page site is unthinkable. A generated page costs almost nothing after the first, so the interesting question stops being 'how many pages can we afford' and becomes 'what does our visitor actually want to look up'.
The hard part is never the pages — it's the data. Most valuable datasets are public but scattered: one register knows who the companies are, another what they earn, a third whether they hold a licence, and their own websites hold the prices. Joining those is the whole job, and it's what we're good at.
What a data-driven portal gives you
- ✓Every record its own page — hundreds of pages that each answer one real search
- ✓Content generated from data, so a change in the source updates every page at once
- ✓A data layer we build: open registers, APIs, or a crawler that collects what nobody has joined up
- ✓Structured data (JSON-LD) on every page type, so Google and AI can read the whole set
- ✓Static output by default — hundreds of pages that still load instantly and can't be hacked
- ✓Search, filters and geography pages where they earn their keep
- ✓A refresh routine, so the data doesn't quietly rot six months after launch
- ✓Full ownership: the generator, the data and the domain are yours
How a portal gets built
Before design, we answer one question: where does the data come from and is it actually public? That decides whether the project is possible at all.
A narrow version goes live early — real pages, real data, indexed. Cheaper to learn from a live slice than from a mockup.
Once the templates hold, the same build produces hundreds of pages. This is the part that costs almost nothing per page.
A monthly data refresh, plus the pages the search data says are missing.
Portal and information system pricing
A portal's price is driven by the data, not the page count. Reading a clean open register is a couple of days' work; crawling four hundred different websites for prices nobody formats the same way is several weeks. So we scope the data layer first and quote it separately — you can even stop there and keep the dataset.
The site itself is comparatively predictable, because generated pages scale: the second hundred pages cost far less than the first ten. That's why a portal can be cheaper than a big hand-built brochure site while doing considerably more.
Data layer only
We build the collector — open-register ingest, an API integration or a crawler — and hand you clean, structured data. Useful on its own.
Portal
Data layer plus the generated site: record pages, hubs, structured data, sitemap, search. The full minuhambaarst.ee pattern.
Data care
Monthly refresh, monitoring and hosting, so the portal keeps telling the truth.
Smaller tasks billed hourly at €70–110 · prices excl. VAT.
Asked about portals
An ordinary site has pages someone wrote. A portal has pages a program generated from data — so 600 of them cost roughly what 10 hand-built pages would, and correcting the source corrects all of them at once.
Usually a public register, an API, or a crawler we write. Estonia publishes a great deal for free — the business register, financial statements, professional certificates, tax-debt lists. The work is rarely finding data; it's joining sources nobody has joined before.
Yes, our own. minuhambaarst.ee lists every dental clinic in Estonia with published prices collected from the clinics' own price lists — 650+ pages, live, with real search traffic. hammasmatka.fi is its Finnish-language sister comparing Tallinn prices against Finnish ones. We built both on this exact stack, which is why we can quote it honestly.
Public data, collected politely, is fine — we respect robots.txt, we rate-limit, and we cite the source on every page. Personal data and login-walled content are a different matter and we'll say no. If a project needs a lawyer's opinion, we'll say that too.
Yes. The same machinery — ingest, structure, generate — makes internal catalogues, price lists and dashboards. Not everything worth building needs to be public.
It will. That's what data care covers: the collector is written to fail loudly rather than silently publish nonsense, and a monthly refresh catches drift.
Automation & web scraping · SEO & website optimization · Website design & development
Got a dataset or an idea for one?
Tell us what the data is and what the pages should answer. If it isn't feasible, you'll hear that in the first reply — for free.