Sourcing

What Is Sourcing? The Complete Guide for Recruiters in 2026

What sourcing is and why it decides whether you fill a role. Techniques, channels, Boolean searches, metrics, AI and GDPR good practice for recruiters and headhunters.

·18 min·The HeyTalent Team · Recruiters & Product
Sourcing

What Is Sourcing? The Complete Guide for Recruiters in 2026

Sourcing in recruitment is the active search, qualification and contacting of talent before you depend on inbound applications. In Spain, where 75% of companies reported difficulty finding the profiles they needed in 2025, waiting for CVs to arrive usually leaves the most interesting candidates out of the process. If you're trying to understand what sourcing is and when to use it, the practical answer is simple: sourcing finds people and opens the conversation, recruiting turns that interest into a hire.

The scene repeats itself in agencies, staffing firms and in-house teams. An engineering role comes in with demanding requirements, the job ad goes live, the ATS starts filling up with poorly matched applications, and the client asks why there still isn't a solid shortlist. Meanwhile, the people who actually fit are already employed and have no reason to go looking for that ad.

Sourcing changes the order of the process. First you define the profile, then you locate people who meet the criteria, verify their details and reach out with a relevant message. After that, the recruiter decides who has enough interest, availability and fit to move forward.

What sourcing is in recruitment, and why it matters in Spain

In operational terms, sourcing is the proactive identification of candidates who could be right for a role, including those who aren't visibly job-hunting. It isn't just looking up names on LinkedIn. It involves translating the client's brief into observable criteria, building queries, comparing career paths, enriching contact data and starting a first conversation.

The definition of a vacancy used by Spain's public employment service, SEPE, helps frame the problem: a paid position already created, or about to become free, for which the company is actively looking for an external candidate and intends to fill immediately or within a set period. That logic describes unmet demand. Posting the role is only one way of making it visible, not a complete strategy for filling it.

In the first quarter of 2023, SEPE estimated 149,645 open vacancies in Spain, 15,657 more than in the same quarter a year earlier. Madrid, Catalonia and Andalusia accounted for 56% of all vacancies, a concentration that forces you to segment searches by region, sector and qualification level.

Sourcing and recruiting do different jobs

Traditional recruiting usually starts once the role is already open. The recruiter posts, receives CVs, filters, interviews, presents candidates and shepherds the offer. The sourcer works before and around that cycle: building the talent universe, discovering profiles that haven't applied and delivering qualified conversations.

Think of an engineering role in Madrid. A job ad can attract active candidates, but a search by stack, project type, location and seniority also lets you locate engineers who are working at competitors and would only consider a move for a specific technical challenge. That's the difference between waiting for visible demand and building your own pipeline.

Rule of thumb: if the role demands an unusual combination of skills, location and seniority, start with sourcing before you post, not after.

Spain's skills mismatch reinforces the need. In 2025, the problem reached 84% in transport, logistics and automotive and 80% in energy and utilities, according to figures compiled by RRHH Press on the difficulty of finding talent in Spain. For an agency, this hits the margin directly: every extra day consumes hours of search, coordination and follow-up.

Four things matter in the decisions that follow: choosing the right channel, writing useful Boolean searches, automating without losing judgement, and measuring the outcome from first contact through to hire. The goal isn't to accumulate profiles. It's to fill difficult roles with less unproductive work.

The difference between sourcing and recruiting

The most useful analogy is the detective and the firefighter. The detective investigates before the alarm goes off: understanding context, following leads, cross-checking information and finding people who aren't making noise in the market. The firefighter acts once the alarm is already ringing: taking the role, organising the response and driving the process to a close.

Both roles are necessary, but mixing them creates the wrong expectations. A sourcer shouldn't just download profiles without checking them. A recruiter shouldn't spend all their time on manual research when interviews, negotiation and client coordination are waiting.

The four tasks that belong to sourcing

  • Identification: locating profiles by job title, skills, companies, location and career path.
  • Initial qualification: checking whether the experience, the stack, the level of responsibility and the professional context fit.
  • Enrichment: completing email, phone or other professional data where it's legitimate and necessary in order to make contact.
  • First contact: writing and sending an approach that explains why the opportunity might be relevant.

The recruiting cycle starts afterwards, although the boundaries shift with team size. It normally covers CV review, interviews, client presentation, expectation management, offer and hire. The sourcer delivers context and opportunity. The recruiter turns that interest into a decision.

Comparative infographic explaining the fundamental differences between sourcing and recruiting in human resources.

A flow you can copy today

  1. Define the real job title. Decide whether you're looking for "Backend Engineer", "Python Developer" or both.
  2. Separate must-haves from nice-to-haves. Don't mix essential experience with the hiring manager's preferences.
  3. Build the search. Add technologies, location, comparable companies and exclusions.
  4. Validate a sample. Review profiles before extracting or contacting at volume.
  5. Send the first message. Personalise the reason for reaching out and record the reply.

Outsourcing makes sense when an agency needs variable capacity, is working a highly specialised role, or lacks internal tools and processes. Building it in-house tends to pay off when hiring volume is stable, there's deep business knowledge and the team can maintain discipline around data and follow-up. A freelancer can cover the middle ground, as long as you agree on qualification criteria, ownership of the information and delivery format.

Sourcing channels that work in the Spanish market

No single channel solves a difficult role on its own. The choice depends on three questions: are there enough up-to-date profiles? Can I get to usable contact data? How much does each conversation started cost me?

LinkedIn Recruiter offers professional context and filters that many teams already know. It works well for mapping companies, spotting moves and locating passive profiles, but the cost and the dependence on a single platform can weigh on an agency running many searches. It also demands manual work if the team has to extract, verify and contact every profile one by one.

Boolean searches on Google are useful for discovering profiles and pages that don't surface easily in a native search. They let you combine job titles, technologies, cities and exclusions, although the information may be out of date and validation takes time. It's a discovery channel, not a guarantee of contact.

Public channels and communities

Spain's Empléate portal and the public SEPE infrastructure show how broad and fragmented the market is. In 2026, Empléate listed more than 10,920 active job ads across Spain, while SEPE was serving 28,639 people a day, operating 712 offices and receiving 1,990,782 monthly visitors to its website. The value for a recruiter isn't in replacing private channels, but in adding regional and sector signals to the research.

Tech communities, industry events and professional groups bring context that never shows up in a database. An engineering meetup can reveal specialists who never answer generic messages, and a technical community can help you understand which alternative names the market uses for a given skill.

AI sourcing platforms bring search, filtering, enrichment and outreach into a shorter flow. Their advantage shows up when a team needs to repeat the process, preserve criteria and cut operational tasks — although automation won't fix a badly defined brief.

For a hard-to-fill role, I'd combine two or three channels. For example, LinkedIn to map target companies, Boolean search to widen discovery, and an enriched database to prioritise who to contact. The winning channel isn't the one that produces the most profiles, but the one that produces the most qualified conversations at a reasonable cost with reasonable traceability. You can go deeper here with a practical guide to people search engines.

Boolean searches and filters that speed up the pipeline

Boolean search remains the central sourcing technique because it forces you to turn an ambiguous description into search logic. The five basic operators are enough to get started:

  • AND requires two concepts to appear, as in Python AND Django.
  • OR allows variants, as in ("Python Developer" OR "Backend Engineer").
  • NOT removes noise, for example NOT internship.
  • Quotation marks search for an exact phrase, such as "machine learning".
  • Parentheses group alternatives and control the logic of the query.

The point isn't to write one enormous string. It's to build blocks you can add or remove depending on the volume and quality of the results.

Three templates ready to adapt

For experienced Python developers in Madrid:

("Python Developer" OR "Backend Engineer" OR "Software Engineer") AND Python AND (Django OR FastAPI) AND Madrid NOT junior NOT internship

For bilingual digital marketing:

("Digital Marketing Manager" OR "Performance Marketing Manager" OR "Growth Marketing") AND (English OR bilingüe OR bilingual) AND (SEO OR SEM OR paid media) AND España

To exclude profiles you don't want to prioritise in a campaign:

("Sales Manager" OR "Account Executive" OR "Business Development Manager") AND SaaS NOT "open to work" NOT freelance NOT "no disponible"

These searches don't prove on their own that someone has a given amount of experience or would agree to relocate. They exist to create a first list. After that you have to check dates, the real scope of the role, continuity of the career path and signals of availability.

Add filters without destroying your volume

Start with job title and skill. Add location only if it's genuinely mandatory. Then layer on companies, sector or previous company size. If you apply every filter from the start, you'll eliminate valid profiles that describe their work using different terminology.

AI can help prioritise variables that are hard to read consistently, such as an estimate of professional seniority, English level or experience at companies of a certain size. Don't treat those variables as hard facts. Use them to order the human review, and document what "strong fit" means in each search.

Quality test: a useful query isn't the most sophisticated one. It's the one a second recruiter can understand, modify and repeat without asking you what you were trying to find.

Save every search alongside the brief, the date, the channel and the reason for the exclusions. That way you can compare results and stop every team member from starting from zero.

How AI and HeyTalent scale the sourcing process

AI adds value when it removes repetitive work without taking over the decision. In sourcing, that means shortening the distance between the brief and first contact, not sending messages indiscriminately.

A professional flow starts with a Boolean query and clear filters: job title, keywords, location, years of experience and company size. From there, the system can help extract profiles, group them by fit, spot relevant signals and complete contact data. The recruiter still validates whether the person fits the client's context and whether the message makes sense.

Screenshot from https://www.heytalent.app

Email and phone enrichment removes one of the slowest tasks in manual sourcing, but it also demands control. You need to know where the data came from, apply minimisation and avoid reaching out through channels that don't match the professional purpose. AI can estimate seniority, detect an advanced level of English or classify previous experience, but those outputs should work as prioritisation signals, not as a substitute for review.

An automated sequence can include a connection request, an initial note and personalised follow-ups. Personalisation doesn't mean auto-inserting the company name. It has to mention a concrete reason for the contact, such as the type of product, the technical responsibility or the commercial scope of the role.

To compare alternatives, separate capacity from cost. HeyTalent's plans are priced in euros and run from freelancers at 2,000 candidates a month up to teams at 14,000 candidates a month, according to the product information provided by the vendor. The operational value depends on how you control credits, criteria, replies and handover to the ATS.

Capability HeyTalent, team plan LinkedIn Recruiter
Search Boolean queries and filters by job title, keywords, location, experience and company size Native search inside LinkedIn
Contact Email and phone enrichment, plus automated sequences Contact through LinkedIn's own features
AI Customisable variables for prioritising profiles Filters and analytics features depending on the product purchased
Management Sits ahead of whatever ATS you already use, without replacing it Requires connecting the result to the team's internal flow
Cost Plans in euros with credit control and defined capacity Price subject to the plan purchased and its commercial terms

The important change is one of process. The ATS — Teamtailor, Viterbit or Workable, say — remains where you manage applications and stages. The sourcing tool works before that, feeding the system with better-filtered contacts and fewer manual tasks.

The market is also moving towards more measurable hiring. LinkedIn's product update for hiring adds analytics signals such as AI message performance, candidate source, response rates and time to accept. For an agency, this makes it possible to compare not only how many profiles each channel finds, but which channel generates conversations that progress.

Before automating anything, define three controls: which profiles go in, which messages go out and when a sequence stops. Without those limits, AI only accelerates the noise.

Key metrics for knowing whether your sourcing works

Sourcing without metrics turns into an argument about opinions. The minimum dashboard has to follow the whole journey, from profile found to hire, because a high response rate doesn't make up for poor qualification.

These are the metrics I'd review weekly:

  • Candidates contacted: measures real activity, separating new profiles from follow-ups.
  • Response rate: tells you whether the channel, the subject line and the message start conversations.
  • Qualification rate: shows what share of replies meets the minimum criteria.
  • Time to first contact: reveals whether the team acts while the role is still a priority.
  • Time-to-fill: connects sourcing to the outcome the client cares about.
  • Cost per hire: takes in tools, internal hours, vendors and fees.

In Spain, the average cost per hire sits between €6,000 and €10,000, the average time-to-fill is 32 days and agency fees run between 12% and 18% of salary, according to Market Research Future's reference on the Spanish AI recruitment market. Don't use those figures as a promise of results. Use them to understand why every hour spent on repetitive tasks eats into the margin.

Infographic on the key metrics for measuring success and efficiency in recruitment strategies.

The one-week dashboard

In a spreadsheet or in your ATS, create one row per campaign and record channel, query, profiles reviewed, messages sent, replies, qualified candidates, interviews and hires. Add the time invested and the associated cost. Don't mix campaigns for junior profiles with executive roles, because their cycles and messaging aren't comparable.

The actionable number is the one connecting two stages. If you get replies but few qualified profiles, targeting is off. If you have qualified candidates but no interviews, the problem may be how the opportunity is explained or how aligned you are with the client. If interviews progress but no offers land, the problem sits further down.

Don't set generic numeric targets without your own history. First establish a baseline per channel and role type. Then decide what to improve: list quality, speed, personalisation or conversion. To go deeper into dashboard design, see this guide to recruitment KPIs.

Legal risk, GDPR and good practice in outreach

Sourcing isn't spam, and the GDPR doesn't ban contacting professionals. What it requires is that you know why you're processing their data, what information you're using, how you obtained it and how they can exercise their rights. In Spain and across Europe, legitimate interest can be a legal basis in certain contexts, but you have to assess it, document it and balance it against the individual's rights.

Transparency matters from the first message. Explain who you are, why you're reaching out, how you identified the profile and how they can object to future communications. Use only the data needed for the professional purpose, and keep reasonable traceability of the source.

Checklist of good practices for running outreach campaigns ethically and professionally.

Checklist before launching a campaign

  • Legal basis: document the legitimate interest assessment or whichever basis applies.
  • Personalisation: tie the message to the profile's real experience and avoid impersonal templates.
  • Objection: include a clear, simple way to stop receiving contact.
  • Source: record where each piece of data came from and when it entered the process.
  • Minimisation: don't store information you don't need in order to assess or contact.
  • Verification: check emails and phone numbers through reliable services and review the results before using them.
  • Limited automation: set pauses, exclusions and stop rules.

Unauthorised mass scraping and unverifiable databases are expensive shortcuts. They can produce incorrect data, duplicates, complaints and a bad reputation for the agency. A sourcing platform with enrichment features can reduce the work, but it doesn't remove the recruiter's responsibility for how the information is used.

AI doesn't replace the recruiter either. It classifies, summarises and prioritises. It doesn't fully understand a team culture, a delicate negotiation or the real reason someone would agree to change employers. This guide to GDPR-compliant sourcing is a useful reference for designing a tidier process.

AI-assisted hiring is growing, but trust remains an operational limit. A survey of 200 senior professionals at large Spanish companies found that 71% were already in advanced stages of adoption, while lack of trust in AI appeared as an obstacle for 37%, followed by talent shortages at 20% and data quality at 19%, according to published data on AI adoption in HR in Spain. Compliance doesn't slow sourcing down. It gives it verifiable limits.

Your 30-day plan for sourcing that scales

Week 1: pick three typical roles from your portfolio. Define must-have requirements, alternative job titles, target companies, locations and exclusions. Save a base Boolean search for each one.

Week 2: set up the sourcing flow, connect the tool you're going to use and create consistent fields for source, status, qualification and consent. Extract 1,000 profiles per role only if that volume answers to a review strategy rather than to accumulating records. Enrich the contacts you're genuinely going to work.

Week 3: launch two outreach sequences with clear variations. Change one variable at a time, such as the opening paragraph, the channel or the angle on the opportunity. Record replies, objections, qualified profiles and conversations that move into the ATS.

Week 4: review performance by channel, cut the messages that generate no interest and document the process so another recruiter can repeat it. Leave the information organised in Teamtailor, Viterbit, Workable or whichever ATS your team uses.

Tech hiring demands particular discipline. AMETIC estimates that Spain's digital sector will need to bring in between 120,000 and 200,000 qualified professionals before 2027, a projection collected in an academic paper available on UVaDOC. If you're looking to save time and money against LinkedIn Recruiter, compare the whole process, from search through to contact, not just the licence price.

HeyTalent offers plans in euros from 2,000 to 14,000 candidates a month, credit control, email and phone enrichment, AI filters and automated outreach. You can try the flow and judge whether it fits your portfolio at HeyTalent, without tying yourself to any particular ATS.

Join the new era of sourcing

Book a call today and start saving time.

Book a demo