Maintain search historySuppose you want to build a job board around a particular market. Maybe you care about:
- UK fintech companies
- Climate-tech startups across Europe
- Graduate employers
- Cybersecurity companies
- Employers selected by your customers
You already know the companies you want to cover. The next question is:
How do you turn jobs from those companies into a useful job-search product?
That involves a few decisions: which employers belong in the board, how job data reaches your application, how users search it, and how much of the underlying infrastructure you want to operate yourself.
↗Related guideHow to Crawl Jobs From Company Career Pages→Start with the market, then choose the companies
A focused job board usually starts with a market, not with the biggest possible job feed. The company list becomes part of the product.
That gives the board a clear boundary, and the difference matters. If someone comes to a climate-tech job board, they expect the jobs to come from companies that genuinely belong in that market.
You also don't need thousands of employers at launch. A carefully chosen set of 100 or 200 companies can be much more useful than a huge collection with little focus.
Design search around the audience
Once the company universe is clear, think about how people will actually look for jobs inside it. Different audiences care about different things.
Graduate board
- Experience level
- Internships
- Location
- Closing dates
Specialist recruitment
- Salary
- Contract type
- Employer
- Distance from candidate
Cybersecurity board
- Specialism
- Seniority
- Remote work
- Clearance requirements
Don't expose every field available in the data. Use the filters that help your audience make a decision.
Decide how the jobs reach your product
There are two common approaches.
Use a Job Search API
SimplerYour product sends the user's search to a job-search service and receives matching jobs back. You don't need to operate your own search index. Works well when your focus is the experience you're building for the user.
Keep your own copy
More controlBring the job dataset into your own infrastructure. More control if you want your own search engine, ranking logic, analytics or data pipelines.
The decision isn't simply API versus export. It is:
Which parts of the job-search infrastructure do you actually want to own?
Make the search feel natural
People don't always use the same language as employers. Someone might search backend developer while an employer calls the role Senior Platform Engineer. A useful job search shouldn't depend entirely on exact title matching.
The same applies to location. Someone searching around Cambridge probably doesn't only mean jobs whose location field contains the exact word "Cambridge". They may also be interested in nearby places such as Ely, Royston or Newmarket.
But keep the visible experience simple. A handful of useful filters is usually better than presenting every possible field at once.
If you use a Job Search API, the search service handles relevance and ranking. If you maintain your own search index from exported jobs, relevance and ranking become part of your own application.
Help users decide quickly
Search results don't need to show everything known about a job. They need to help somebody answer:
Is this worth opening?
Senior Software Engineer
The job-detail page can then provide the full description and other structured information.
You don't necessarily need to build an application system yourself. The job board can concentrate on discovery and send the candidate to the employer when they're ready to apply.
Grow the board around the market
A good company-first job board doesn't need to begin with perfect coverage. Start with the companies that matter most, then expand deliberately.
Growth should improve the usefulness of the board, not simply increase the number of records. The company universe can evolve alongside the product.
Build on a company-first job index
With Recuity, you choose the companies that belong in your job universe. Recuity discovers and structures their jobs and makes the resulting index available to your product.
Use Recuity as the search engine behind your own website, job board or application. Your users search through your interface, while Recuity handles the job search behind it.
If you prefer to operate your own search infrastructure, export the tracked jobs into your own database, search engine or downstream systems.
Use JSONL when you want to own that layer yourself.Repeated searches and job retrievals through the API don't consume the indexed-job allowance, subject to your plan's API limits.