57 factual definitions so you speak the same language as your acquisition teams, your agencies and your software vendors. Three families: acquisition and distribution, qualification and data, market and roles.
The process by which a school or training body selects and enrols its future learners: application, interview, decision, confirmation. The performance of an admissions team is measured by the number of complete applications received and the cost per application, not by clicks. Campaigns that feed admissions are timed on the peaks of the academic calendar, the logic behind the pricing designed for schools and training providers.
A site that automatically collects job ads published elsewhere and republishes them in a single search engine. The aggregator produces no ad of its own: it indexes those of companies and job boards. Its traffic depends on search rankings and on how fresh its feed is. Publishers who want to turn that traffic into applications can sponsor their clients' ads through the offer designed for job boards and aggregators.
A training format that alternates periods in a company and periods in a training centre, under an apprenticeship or professionalisation contract. It involves three parties: the learner, the employer and the training body. Recruiting for work-study therefore combines an academic admissions logic with a classic hiring logic, with two calendars to keep at the same time.
A work-study employment contract aimed primarily at young people, leading to a diploma or professional qualification. It is supervised by an apprenticeship tutor in the company and partly funded by skills operators. For an apprentice training centre, every unfilled place is lost funding, which makes filling seats as critical as teaching quality.
Applicant tracking software that centralises job ads, the applications received and the steps of the hiring process. An ATS is a management tool: it organises what comes in, it does not generate inbound flow. It sits downstream of acquisition and qualification. The AI talent qualification agent connects to 100+ ATS to send back profiles that are already scored.
A group of people sharing the same occupation, level of experience or set of skills, rebuilt from browsing signals and profile data. It is the basic targeting unit of a talent acquisition campaign. Seeqle works with 5,100 of them, built on 400,000+ targeting criteria and fed by the Seeqle Intelligence data engine.
A systematic deviation in a model's results that disadvantages a group of candidates for reasons unrelated to their actual skills. It most often comes from training data that reproduces past human decisions. Controlling it means documenting the criteria used, measuring the gaps and keeping a final human decision. It is the first thing to check before deploying an automated pre-qualification set-up.
The amount committed to distributing job ads as targeted advertising, separate from the platform subscription. It is managed per campaign and reallocated along the way towards the channels that produce applications. At Seeqle, the base plan is €300 per month and the sponsoring budget stays in the client's hands: the recruiting pricing model clearly separates the two lines.
An external provider mandated by a company to identify, approach and present candidates for a given position. It is paid on success or on a fixed fee, and often handles several mandates in parallel. Its main constraint is the time it takes to present a shortlist. Harry Hope uses Seeqle on its own assignments, within the framework described in the programme for agencies and recruitment firms.
A distribution campaign kept running continuously rather than launched in bursts each time a position opens. It maintains a constant level of visibility and feeds a talent pool before the need is even expressed. Its point is to smooth costs and avoid bidding wars at peak times. It works well on long-attention channels such as audio and streaming platforms.
Restricting the distribution of an ad to a precise geographic area: a radius around a site, an employment area, a department, a country. It avoids paying for impressions outside candidates' realistic commuting zone. For a network with many outlets, it allows one campaign per location. It is one of the basic settings of advertising distribution.
The full cost of a completed hire: distribution, tools, team time, external fees, divided by the number of hires made. It differs from cost per application, which measures only one step of the journey. A high cost per hire more often signals a qualification problem than a volume problem. What a hire really costs in France, backed by public sources
Distribution budget divided by the number of applications received over a period. It is the most readable efficiency indicator of an acquisition campaign, provided it is cross-checked with the qualification rate. A drop in cost per application paired with a drop in quality is not a gain. The published customer cases give the orders of magnitude observed by sector.
A media buying model in which the advertiser pays for a predefined action rather than for mere exposure: form submitted, account created, application filed. It shifts part of the performance risk to the ad network. In recruiting, the chosen action is almost always the completed application.
A media buying model in which the advertiser pays for each click on its ad, whatever happens next. It is useful to compare the appeal of two creatives, but insufficient to judge a recruiting campaign: a click is not an application. Levels vary widely by channel, notably on professional inventories such as LinkedIn.
A relationship database that keeps the history of exchanges with candidates, including those who were not selected, so they can be reactivated later. It differs from the ATS, which is centred on tracking a process in progress. Its usefulness depends entirely on data freshness and on the ability to segment the pool.
Automatic extraction of the structured information in a CV: identity, experience, degrees, skills, dates. Parsing turns a free-form document into comparable data. Its quality depends on the source format and drops on highly graphic CVs. It is a technical building block, not an assessment: it orders information, it does not judge fit for the role.
All the personal information collected during a hiring process: contact details, background, exchanges, assessments. It falls under the GDPR and requires a legal basis, a defined retention period and clear information to the person concerned. Data enriched from a public source remains personal data and follows the same rules.
The perception a candidate keeps of their whole journey, from discovering the ad to the final answer. It comes down to measurable details: form length, response time, clarity of the process. A poor experience is paid for in abandoned applications and negative public reviews. The free HR templates and checklists let you audit each step.
A grouping of French public secondary schools that pool their resources for adult continuing education. Each GRETA reports to a regional education authority and covers a defined territory. Its acquisition challenges resemble those of a private school: filling sessions with fixed dates. GRETA Nord Alsace works on this need with Seeqle, within the offer for schools and training providers.
Technical skills that can be verified and most often certified: mastery of a software, licence, language, degree, driving permit. They are assessed through a test, a qualification or documented experience. They form the eligibility filter of an application, rarely the final decision criterion.
An artificial intelligence system in which every result can be traced back to the criteria that produced it. In recruiting, it means being able to say why a profile gets a given Score, and on which data. It is a practical condition for acceptance by HR teams and for compliance in case of dispute. The Seeqle Intelligence data infrastructure exposes the criteria behind each scoring.
The display of an ad on a user's screen, counted once per delivery. It is the unit of measurement of advertising reach, not of interest generated. A high volume of impressions with no clicks or applications points to a message or targeting problem, not a lack of budget.
The temporary provision of an employee by an agency to a client company, for an assignment of defined length. The model requires filling positions fast, often within days, in high-demand occupations. The constraint is not one-off sourcing but the continuous feeding of an available pool, which is what the offer for staffing agencies is built for.
A network of job sites, specialised media and partner apps on which the same ad is distributed simultaneously. It widens reach without multiplying contracts or interfaces. Its value depends on the quality of the inventories more than on their number. The Seeqle job network details its composition.
A site specialised in publishing job ads, visited by candidates who are actively looking. A job board captures an intent already expressed; it does not create discovery. It remains effective for high-demand occupations and becomes expensive for rare profiles. Seeqle's partner job boards use it to sponsor their own clients' ads.
An identified contact showing interest in a programme: brochure request, open-day registration, pre-registration form. It is the education-side equivalent of a sales lead. Its value depends on how far it is from a complete application. The same capture mechanics apply to B2B lead generation.
The image a company projects as an employer, to its own staff as well as to the market. It is built over time and measured in spontaneous awareness, direct traffic to job ads and offer acceptance rate. A weak employer brand is paid for in cost per application. Visibility campaigns work on this asset outside hiring peaks.
The mechanism that maps the requirements of a job to the characteristics of a profile. It compares weighted criteria: skills, experience, mobility, availability. Matching is the mechanism; the Score is the numerical result it produces. Usable matching makes its criteria explicit instead of delivering an opaque ranking.
An occupation for which the number of open positions durably exceeds the number of available candidates in a given area. Shortage is local before it is national: the same job can be hard to fill in one region and not in another. For these positions, a classic ad is no longer enough. The sector-by-sector approach details the channels that work, occupation by occupation.
Simultaneous publication of the same ad on several sites from a single interface, with centralised return of applications. It removes re-entry and harmonises ad content. It solves a distribution problem, not an audience problem: if the chosen sites do not reach the right profiles, volume stays low. Multiposting of ads across the Job Network
The equivalent of the employer brand for a school, university or training body: what the public knows about the institution before even comparing programmes. It drives the volume of spontaneous applications and selectivity. It is built on the channels used by secondary-school and university students, notably TikTok short-form video.
The integration period of a new hire, from contract signature to the first months in the role. Poorly framed onboarding is one of the most frequent causes of departure within the first year. It is prepared before arrival, not on day one. The HR guides and white papers detail the steps to formalise.
A French state-approved skills operator responsible for funding apprenticeship and vocational training for a given industry branch. It conditions part of the budget of apprentice training centres and training providers. Identifying the reference OPCO of a sector is a prerequisite to setting up a work-study offer.
A situation in which the skills sought are durably scarcer than the needs expressed by the market. It is not fixed by publishing more ads: the profiles concerned do not read job ads. The answer lies in widening the pool addressed and shortening processing times.
An assessment step between receiving an application and the interview, which determines whether the profile deserves human time. It relies on explicit criteria and on enriching the available information. Done well, it protects recruiters' time without discarding atypical profiles. AI pre-qualification at Seeqle reaches a 53% qualification rate on average.
A person in a job who is not actively looking but remains open to a relevant opportunity. They visit neither job boards nor aggregators, so a classic ad never reaches them. These profiles account for 80% of talent. Sourcing through targeted distribution exists precisely to reach them where they spend their time.
A technical category covering the automated, real-time auction buying of ad space based on audience criteria. Applied to recruiting, programmatic HR is limited to distributing ads at the best cost of exposure. It is a media-buying foundation, not a complete answer: it does not qualify the applications it generates. Seeqle goes beyond this category by linking the distribution of the Attract Agent to the scoring of the Match Agent on a single data infrastructure. What Seeqle Intelligence does beyond a programmatic ad network
Delivering a message to an audience defined by profile, behaviour and context criteria rather than to a general audience. It makes it possible to reach people who are not looking, on sites unrelated to employment. Seeqle distributes on Facebook, Instagram, TikTok, LinkedIn, Snapchat, YouTube, X, Reddit and Spotify, then on Google Search and Google Display, Indeed and job aggregators, online press and apps, across an inventory of 2M+ sites and apps, through partner advertising inventories and social networks.
A measure of a hire's performance after the start date: holding the position, six-month review, still on the payroll after one year. It is the only indicator that links acquisition to operational results. It requires collecting information held by managers, not by HR. Defining these indicators takes preliminary framing, the subject of a first conversation with a Seeqle specialist.
The professional in charge of identifying, assessing and selecting candidates for an organisation's needs. The job has shifted from sorting applications to building audiences and running campaigns. How companies use the AI Agents shows how these two activities combine day to day.
The General Data Protection Regulation, applicable to any processing of personal data in the European Union. In recruiting, it requires a declared purpose, a limited retention period, and access and deletion rights for the candidate. It does not prohibit automated scoring, but strictly frames fully automated decisions.
Return on advertising spend transposed to recruiting: the value produced by a campaign divided by the distribution budget committed. The difficulty is defining the value of a hire, which varies by position and by the cost of a departure. Failing that, cost per qualified application serves as the benchmark. The talent attraction diagnostic helps lay these calculation foundations.
Outsourcing all or part of the recruiting process to a provider that operates on behalf of the client company. RPO differs from an agency by its volume logic and its integration into the client's tools. Its margins depend directly on the time spent per file processed. This model is described in the partner offer for agencies and RPOs.
The numerical result assigned to a profile at the end of matching, on a scale defined in advance. It serves to sort and prioritise, never to decide on its own. A Score only makes sense when it comes with the detail of the criteria that produced it.
An indicator estimating the probability that a contact will complete a process: file an application, enrol in a programme, sign. It relies on behavioural and declarative signals rather than on the profile alone. On the school side, it focuses admissions teams' effort on the applications that will convert; it is included in the education base plan.
The operation that assigns a Score to each profile by applying weighted criteria to enriched data. Scoring makes applications of heterogeneous formats comparable. It is set up per position: the criteria for a maintenance technician are not those for a salesperson. The Match Agent runs this scoring and sends it back to the ATS.
Behavioural and interpersonal skills: communication, autonomy, conflict handling, adaptability. They are inferred from career elements and observable clues, never from a generic personality test. In customer-facing roles they often weigh more than hard skills in on-the-job success.
Actively searching for profiles that match a need, beyond the spontaneous applications received. Manual sourcing relies on exploring databases and networks, profile by profile. Sourcing through targeted distribution reverses the logic: the ad goes to the profile, and the campaign launches in 2 min.
The activity of providing personnel, employed or freelance, for fixed-term assignments. The term covers temp work, umbrella employment and skills subcontracting. A staffing player's performance is measured by its placement rate, hence by the depth of its available pool. The temp and staffing offer is built on this indicator.
The ratio between the number of clicks and the number of impressions of an ad. It measures the appeal of the message and the visual, not the relevance of the targeting. A high click-through rate followed by a low application rate signals a promise the landing page fails to keep.
The share of visitors who land on a job ad and actually apply. It depends mostly on form length, ad clarity and reading comfort on mobile. It is the cheapest lever to fix on an underperforming campaign. The Seeqle HR blog documents the optimisations that matter most.
The share of applications received that actually match the position's criteria. It sets raw volume against useful volume and determines the sorting workload of the teams. A low rate means most files are read for nothing. Seeqle campaigns reach a 53% qualification rate on average.
The ratio between the number of learners enrolled and the number of places open on a session. Below a threshold, the session is cancelled while fixed costs remain committed. It is the economic survival indicator of a training body. Campaigns for schools and training centres are run against this objective.
The time elapsed between a position opening and the selected candidate accepting the offer. It adds up distribution time, sorting time and decision time. It is the first indicator operational directors look at, because a vacant position has an immediate cost. The Attract Agent acts on the first of these three steps.
Staff rotation rate over a period, calculated from departures relative to average headcount. High turnover turns recruiting into a permanent race and damages the employer brand. Its causes are rarely linked to sourcing: they lie in integration and management. The HR webinars and podcasts regularly cover this topic.
All the profiles already known to an organisation: past applications, sourced contacts, event attendees. A pool only has value if it is segmented, up to date and reactivable within minutes. Without prior qualification, it becomes a dead database. The AI qualification agent keeps this pool scored as applications come in.
This glossary is kept up to date as the recruiting, education and acquisition-technology markets evolve.