Something I’ve learned while building with company data:
The hard part is usually not displaying the data.
The hard part is finding it, cleaning it, checking it, structuring it, and deciding how much to trust it.
That is what led me to build FastBusiness API.
At first, the problem seemed simple. I wanted to take a company or business name and return a useful business profile. Things like the website, industry, sector, headquarters, business type, identifiers where available, a short description, source links, and a confidence score.
On the surface, that sounds like a basic enrichment problem.
But once I started building it properly, I realised the real challenge is not just getting data. It is turning messy public information into something consistent enough that another developer could actually use inside a product.
A business name by itself is not always enough.
Sometimes there are multiple companies with similar names. Sometimes the website is unclear. Sometimes the business has a trading name that differs from the legal name. Sometimes the company is global, but the user is looking for the Australian entity. Sometimes public identifiers are available, sometimes they are not. Sometimes the first search result looks right but is actually a directory page, a reseller, a subsidiary, or an unrelated company.
That makes the API design more interesting than I expected.
It is not enough to return:
“Here is the company.”
The API needs to communicate:
“Here is the company I think matches your request, here is the structured profile, here is where the data came from, and here is how confident the system is.”
That is the part I have been focusing on.
A few things I am thinking about while improving it:
- matching logic for similar business names
- optional website and country inputs to reduce ambiguity
- consistent response structures across successful and failed lookups
- confidence scores that actually mean something
- source links so returned data can be checked
- clear handling of missing fields
- refresh logic for stale business profiles
- pricing that makes sense for small builders, not just enterprise users
- documentation that shows real examples rather than vague marketing
One thing I have realised is that “more data” is not always better.
If an API returns 40 fields but half of them are unreliable, inconsistent, or unclear, that does not help the developer much. In some cases it makes the product harder to trust.
For business/company data, I think the more useful goal is:
return the right fields, in a predictable structure, with enough context to know whether the result should be trusted.
That is why I am trying to treat sources and confidence as core parts of the response, not as extras.
For example, if a CRM, internal dashboard, lead enrichment tool, directory, or due diligence workflow uses company data, the user probably does not just care about the raw answer. They care whether the answer is good enough to rely on.
That changes how I think about the product.
FastBusiness API is not just meant to be a search wrapper. The goal is to sit between messy public information and the clean structure a developer actually wants to work with.
The ideal flow is simple:
A developer sends:
- business name
- optional website
- optional country
The API returns:
- structured business profile
- useful company context
- source links
- confidence score
- predictable fields
- clear errors when the match is weak or missing
That sounds simple, but a lot has to happen behind the scenes to make the response feel boring in a good way.
And honestly, that is what I want from the API.
I want the output to be boring, predictable, and easy to integrate.
The messy part should happen behind the API, not inside the customer’s app.
I am still early, but I have been improving it based on feedback and trying to make the product more useful for real workflows instead of just building what sounds good on a landing page.
The main use cases I am building around are:
- CRMs
- lead enrichment
- business directories
- internal dashboards
- sales tools
- vendor/customer research
- due diligence workflows
- products that need company context without building their own research pipeline
The bigger lesson for me so far is that data products live or die on trust.
If the response is inconsistent, developers will not use it.
If the source is unclear, users will question it.
If the match is ambiguous, the API needs to be honest about that.
If the structure changes too often, integration becomes painful.
So the real work is not just collecting business data.
The real work is making company data usable.
That is what I am trying to solve with FastBusiness API.
Site: https://fastbusinessapi.com
For anyone who has worked with enrichment APIs, company data, CRMs, directories, or internal tools:
What makes business data trustworthy enough for you to use in a real product?
Is it source links, confidence scores, field-level confidence, refresh dates, manual overrides, or something else entirely?
Top comments (0)