01
Two-minute brief
Tell us about your business and pick this service. The listed price anchors every quote: no bait, no mystery ranges.
Concept previews illustrating this service's typical deliverables.
What this is
The service, straight up.
Every industry has its information sitting in plain sight and in completely useless shape. Competitor pricing on forty different websites. Tender notices across nine government portals. Company registrations, licence registers, planning approvals, import records, job ads, product catalogues, association member lists. Any one of them is a browser tab. All of them together, tracked over time, is a genuine strategic asset, and nobody has it because assembling it by hand is a full-time job that no one has ever been given.
This build gives you that asset as software. Collectors run on a schedule against each source, whether it offers a clean API, a downloadable file or nothing but a public web page. Everything lands in one normalised schema, so a company that appears in four sources under four slightly different names becomes one record with four citations. Records are versioned, so you do not just see that a competitor charges $1,450 today, you see that they charged $1,190 in March and exactly when it moved. Then the interface on top does the part most data projects skip entirely: full-text and faceted search, saved views, side-by-side comparison, change alerts, charts, and a one-click export that your analysts can actually use.
Being honest about the hard parts, because they are what the price reflects. The collection is the easy half; the difficulty is entity resolution, deduplication, and knowing when a source has quietly changed shape rather than gone quiet. Every collector is monitored with freshness thresholds and alerts, every field carries provenance back to the source and timestamp it came from, and low-confidence matches are queued for a human rather than silently merged. We scope legality per source before writing a line of code: what is public, what the terms permit, what is licensed and needs paying for, and what we will not touch.
The result is usually the same wherever it lands. Sales knows which accounts moved before the account manager hears it. Pricing stops being a quarterly guess. Product sees a gap in the market as a filtered list rather than an anecdote. And the question "how big is this segment actually" takes four seconds instead of two weeks.
What's included
- Source-by-source scoping: legality, licensing and feasibility first
- Scheduled collectors for APIs, files and public web sources
- One normalised schema with entity resolution and deduplication
- Full history, so every field is versioned and dated
- Search, facets, saved views, comparison and charts
- Provenance on every field, back to source and timestamp
- Freshness monitoring with alerts when a source changes shape
- Role-based access, audit log and API or CSV export
How it works
Fixed price, not fixed blind.
02
Three of the 11
Within 24 hours, three of the 11 vetted developers who deliver this service send written, fixed-scope quotes. Compare and choose.
03
Build, guaranteed
Typical delivery in 8–12 weeks. If a matched developer drops the ball, we step in. Service Guarantee, always.
FAQ
Fair questions.
It depends entirely on the source, which is why the first phase is scoping each one rather than building. Public registers, government open data and licensed feeds are straightforward. Public web pages depend on the site terms, the robots file and what you do with the result. Personal information brings the Privacy Act into it. We categorise every proposed source as clear, needs a licence, or will not touch, and you see that assessment before development starts. We will decline a source rather than expose you.
That is the normal state of affairs, not an edge case, and it is designed for. Every collector has expected volume and freshness thresholds, so a source that silently returns nothing or half as much raises an alert the same day rather than quietly rotting for a quarter. Structural change detection flags when a page still loads but no longer parses. Maintenance for source changes is quoted as an ongoing arrangement, because pretending it is a one-off build would be dishonest.
Entity resolution, and it is the part that takes the real skill. Records are matched on the strong identifiers first, an ABN or ACN or a registration number, then on fuzzy signals like name, address and domain. Confident matches merge automatically and keep all their citations. Anything ambiguous goes into a review queue for a person to decide, because a wrong merge is far more damaging than a duplicate and much harder to notice.
It is your data and your platform. CSV and Excel export from any view, a documented read API for your own systems, and direct database access if you want your analysts in there with their own tools. You own the code and the schema outright. Being locked into a vendor is the problem this is meant to solve, so building a new one would be self-defeating.
Roughly half and half, and the interface half is the one people underestimate. A pile of correct data with no way to interrogate it gets used twice and abandoned. The search, the facets, the saved views and the comparison screens are what turn it from a database into something a non-technical person opens every morning.
Related services
