Search and Match

Traditionally HR search and match is a fairly simple affair which involves indexing the candidate data in a database or search engine, then querying that data with various keywords or phrases.

Whilst this works, mostly, it is both time-consuming and prone to error or missed results.

For example, if you had an opening for a “customer service assistant” then the natural way, using the above, would be to query “customer service assistant”. If you wanted to broaden results then you might query “customer service” but then you’d have to wade through the “customer service managers” and “customer service directors” too. If you were a bit smarter you might also search for a synonym, e.g. “customer service assistant, customer service asst”. But you will miss a great candidate for the role who has “customer svc assistant” on their CV

See where we are going with this? To search for all common variations of “customer service assistant” would require around 10k (yes, ten thousand) synonyms (according to our taxonomy), including abbreviations, typos and role synonyms. Simple text search just doesn’t cut it any more.

But even if you are lucky enough to find some candidates using traditional search, how do you rank them? Typically the highest ranked will be those who have mentioned “customer service assistant” the most times on their profile. Not necessarily that useful.

As for matching jobs to candidates, or vice versa, then almost everything on the market is pretty poor and extremely prone to false positives and true negatives!

A good search and match suite should be able to:

  • Find candidates or jobs using all variations of the roles and skills involved

  • Be able to find a candidate by simply searching using the job text itself, or find a job by using the candidate CV

  • Automatically create a ranked candidate shortlist for a given vacancy

  • Accurately match candidates to vacancies based on important and the higher value skills

  • Match vacancies based on a minimum level of experience or freshness in the role or skills

  • Avoid false positives such as “made tea for the CEO” when searching for a CEO

A poor search and match means that you, or your users, will spend hours trawling through irrelevant results, or that you will miss the perfect candidate or job. A good search and match will save time and money by quickly finding the candidates/jobs you are looking for and quite possibly reactivate the 90% of dormant candidate data sitting around in your database.

The reshufl search and match solution


Used in conjunction with the reshufl parser, the reshufl search and match delivers extremely accurate results with the minimum of effort. Our 6-way search enables you to search vacancies/job orders by query, another vacancy or a CV, and to search CVs by query, a vacancy or another CV.


When searching for “customer service assistant” our search engine doesn’t look for that text. Instead it looks for the reshufl internal concept of “customer service assistant”, and this single concept includes all 10k variations of that role, including abbreviations, synonyms and typos.


Our AI engine looks at a myriad of different tell signs to indicate important terms to be used for matching. The table below shows examples of some of the complexities involved in deciding what’s important.

Semantic Context

Whether a skill is relevant to the work experience role which it is associated with. e.g. the term “java” found as part of a software developer experience description is significant but if found when describing a sales person’s experience then it’s not likely to be. Therefore, its importance to the relevance of the results is downgraded.

Experience Freshness

The freshness of a candidates experience is crucial when matching. The waiting in a restaurant experience of a graduate from 20 years ago is not very relevant to job matching today unless of course the candidate is still in the same role. 

Identifying when each role and skills was used allows our AI engine to intelligently decide which of them is most relevant, in much the same way as a human would.

Current Market Value

Roles and Skills with a higher demand/supply ratio in the current job market are likely to be more valuable and hence become more significant when matching jobs than their lower value counterparts. e.g. future employers of statisticians are likely to value a candidate’s experience with the Statistica package over his experience with Excel experience. As reshufl has extensive roles and skills valuations, these can be used to determine the best matches.

Experience Duration

The length of time a candidate has been carrying out a particular role or using a specific skill is potentially very important. This duration feeds through to our AI matching engine so that it can be intelligently taken into account. 

We can also enforce these requirements to say, for example, that candidates must have at least 5 years’ experience (spread across multiple experiences if necessary) when being matching to a specific job. 

Semantic Relationships

Our AI matching engine looks at:

  • identical in meaning (including typographic variations) e.g. “software developer” and “sw developer”

  • term synonyms e.g. “software developer” and “programmer”

  • related terms e.g. “software developer” and “it consultant programmer”

The resulting huge term expansions allows for significantly higher recall of matching candidates/vacancies so that your business or users don’t miss any opportunities

No more need for long boolean queries and attempts at guessing synonyms, reshufl does all the work. No other search in the industry comes close to ours.

Soft Skills

We identify and ignore where appropriate skills such as “hard working” that are essentially noise and don’t contribute to finding a great match. 

The reshufl search and match is completely configurable, and easy both for you or your users to change key parameters. At its simplest, a user can upload a CV which will then display the best matching jobs from your database, or you can simply paste a vacancy and see a shortlist of candidates immediately.


Multiple operations are supported too, e.g. submit multiple CVs and find the best jobs for each, for a jobs by email process for instance.


reshufl supports uploads of CVs or vacancies via raw upload, file system scan, database connections or Amazon S3 storage. All easily configurable.

Why Choose Us?

Put simply, we are the most accurate search and match available on the market, way ahead of all our rivals, but don’t take our word for it, compare us now to the others.

We don’t say that our matching will replace the recruiter. All jobs are different, as are all candidates, and the successful match of both will almost always come down to factors other than those on a CV, but our matching algorithms will dramatically cut down time taken to shortlist candidates and provide the valuable insights into your data that you need.

We are also agile enough to grow with your business in ways that some of the incumbent players cannot. If you have special or bespoke requirements, then we can move quickly to accommodate you. Please see our strategic partnerships page for more information and get in touch with us for more information.