Workable Applicant Tracking System: A Buyer's Read

Workable made your shortlist, the demo is next week, and somebody has asked you to come back with a recommendation rather than a feature grid. The awkward part is that you are shopping for one thing and being sold another, and the vendor is not hiding it — their own product page opens with the words "your complete HR and hiring platform".

That sentence is the most useful thing on the site for a buyer. You went looking for a Workable applicant tracking system. What is for sale is recruiting, hiring and people management in one integrated platform. Whether that is a bonus or a problem depends entirely on what you already run, and it is the question that decides this purchase — not the pipeline screenshots.

So this piece works through it in the order that matters. What the recruiting half genuinely does well, which is more than the sceptics allow. What sits in the base tier and what is a separate line on the invoice. How the bill is counted now, which is not how most comparison articles still describe it. Where an ATS-first platform collides with the HR system you already have. The three jobs you will still be doing by hand afterwards. And then the six questions to put to their sales team, with a note on what each answer tells you.

What does the Workable ATS include?

Two things in that paragraph tend to surprise people. The first is how much HR functionality sits in the base plan of a product most buyers think of as an ATS. The second is that three features many buyers assume are core — texting candidates, one-way video interviews, and assessments — are not in the entry tier without paying extra.

Distribution is the clearest strength, and it is a real one

If you take one thing from this article about what Workable is good at, take this. The published reach is over two hundred free and premium job boards from a single posting action, alongside a no-code careers page builder that produces a branded, mobile-friendly page with search and filtering.

That combination solves a problem that is genuinely tedious to solve any other way. Posting one role to eight boards by hand is an hour, plus eight sets of credentials, plus the reconciliation afterwards when applications arrive in eight inboxes. Multiply by the number of roles you run in a year and the distribution feature alone can justify a mid-market ATS for a company that hires steadily.

The sourcing side goes further than distribution. The published figure is over 400 million searchable candidate profiles with AI-powered recommendations, which is a different product category bolted onto the same platform: distribution waits for people to come to you, sourcing goes and finds people who are not looking. There is also an autonomous agent product covering sourcing, screening and engagement, available on every plan with an allowance of AI credits.

Three caveats worth carrying into the demo, none of which are objections.

"Free and premium" is doing work in that sentence. Premium boards charge. The platform gives you one place to buy and manage those postings, not free access to paid inventory. Ask which of the boards that matter in your market are free and which are pass-through cost.

Distribution solves volume, not fit. If your problem is nobody applying, more boards helps. If your problem is that the wrong people apply, more boards makes the reading problem worse, and the reading problem is the expensive one. Be honest about which you have before you weight this feature.

A large candidate database is not the same as candidates who want your job. Every sourcing product in the category is built on the same shape of promise, and the useful question is reply rate rather than database size. Ask what proportion of contacted candidates respond, for roles like yours, and treat a refusal to answer as an answer.

The tiers, and the three features that cost extra

As published in August 2026, the entry figures are Standard at $299 a month or $3,588 a year, Premier at $599 a month or $7,188 a year, and Enterprise at $719 a month or $8,628 a year. Those figures move with company size, so treat them as the shape rather than as your quote.

Standard includes candidate sourcing, applicant tracking, hiring plans, an interface in nine languages, and the HR side: an HRIS, onboarding, time-off management and payroll prep. What it does not include is texting, one-way video interviews and assessments — those are add-ons at $89, $109 and $59 a month respectively at that tier. Premier and Enterprise bundle all three in, and add referrals plus a performance and engagement module.

Run that arithmetic before the call, because it decides the tier. Standard plus all three add-ons is $299 plus $257, which is $556 a month — against Premier at $599 with those three included plus referrals and the performance module. If you need all three, Premier is the better deal by a margin small enough that the extra modules are effectively free. If you need one of them, Standard plus that one is clearly cheaper. If you need none of them, Standard is right and the upgrade conversation is a distraction.

So the real question is not which tier. It is which of texting, video screening and assessments you will actually use, and that is answerable in one meeting with the people who run interviews. Most teams discover they want texting, are unsure about video, and are not going to run assessments without somebody owning the validation of them.

The trial is worth using properly: fifteen days with the complete Standard feature set and no credit card. Fifteen days is enough to run one real role end to end, and running one real role tells you more than three demos. Do not spend the trial clicking around — post an actual opening, take actual applications, and do the reading inside the product.

How the bill is counted, and why the old advice is wrong

A large amount of the comparison writing about this product still describes it as priced by open job slots. That was true once and is not the current published model, and the difference changes which companies it suits.

The published model prices by company size bracket — 1 to 20 employees, 21 to 50, and upward — with unlimited active jobs on all plans, subject to fair usage limits.

Three consequences follow, and they favour different companies than the old model did.

Seasonal and burst hiring becomes cheap. A company that opens twenty roles in March and two in August pays the same either way. Under a per-slot model that same pattern was punished exactly when the hiring was hardest. If your hiring is lumpy, this shape is the one you want.

A large company that barely hires now pays for its size. The mirror image. Eight hundred employees and four openings a year puts you in a high bracket for a product you use lightly, and the per-slot model would have been kinder. This is the case where the arithmetic should be run properly before anyone signs.

"Subject to fair usage limits" is a phrase to resolve. Unlimited with a fair usage clause is not unlimited; it is undefined. Ask what the threshold is, what happens when you cross it, and get the answer in the contract. Vendors who have thought about it will answer immediately.

It is worth seeing this shape next to the opposite one. BambooHR bills by headcount and then caps the hiring module by open roles — five on the entry plan — which is the reverse arrangement and suits the reverse company. Neither is wrong. They are aimed at different hiring patterns, and the mistake is buying without knowing which pattern you have.

Where an ATS-first platform collides with an HRIS-first one

Now the part that decides whether this purchase is clean.

Workable ships an HRIS in its base plan: employee profiles, onboarding, time-off management, payroll prep, and, at the higher tiers, performance and engagement. If you have no HR system, that is genuinely valuable — one purchase covers hiring and the first year of employment, with no integration in the middle.

If you already run an HR system, you have just bought a second one. Not deliberately, and not because anybody misled you, but because the platform is sold as a bundle and the HR half is not optional in the way the add-ons are.

The practical resolution is one decision made explicitly and written down: which system is the source of truth for employees. Everything else follows.

If Workable is the source of truth, your existing HR system becomes a downstream consumer or gets retired. That is a real migration with a real cost, and it is a legitimate choice — particularly if the incumbent is a spreadsheet and a shared drive.

If your existing system stays the source of truth, then you are buying Workable for hiring and deliberately not using its HR half, and the handover point is the moment somebody accepts an offer. This is the cleaner arrangement and the one most companies with an existing HRIS should choose. Say so out loud in the implementation meeting, because otherwise somebody will helpfully start entering employee data in both places and you will discover it a year later during an audit.

What you must not do is let both systems hold employee records with a two-way sync between them. They will disagree, silently, and you will find out from a payroll error. Pick a direction and keep it one-way.

Screening and scoring: how far it goes

The published capability set covers custom hiring pipelines, a talent CRM database, AI-powered candidate recommendations, and — as a paid add-on at the base tier — science-backed cognitive and personality tests. Reporting includes pre-built reports and a custom report builder covering pipeline, time-to-hire and diversity metrics.

That is a solid mid-market set, and there are two things worth understanding about the shape of it.

The first is that recommendation and ranking are different from evidence. A ranked list tells you the order somebody's model put the candidates in. Whether it tells you why, at the level of the specific line in the CV that produced the score, is the question to ask in the demo — and it is the question that determines whether you can defend a rejection six months later. Ask them to show you a scored candidate and then ask what the score was based on. The answer separates the products more than any feature list does. If you want the full version of that argument before you sit down, what AI screening tools can and cannot substantiate is the piece to read first.

The second is that assessments are a commitment, not a feature. Cognitive and personality tests have to be chosen for a specific role, validated against something, and applied consistently, or they add legal exposure rather than signal. Buying the add-on is the easy part. Somebody has to own it. If nobody in your organisation is going to own it, do not buy it, and the $59 a month is not the reason.

Parsing quality sits underneath all of this and nobody demos it. A screening tool that reads CVs badly produces confident scores from misread data, and the failure is invisible because the score looks the same either way. How parsing actually works and where it breaks is worth twenty minutes before you evaluate any scoring claim, from any vendor.

The careers page, and the part of it that decides your application rate

The no-code careers page builder is listed as a feature and treated by most buyers as a checkbox. It is worth more attention than that, because the careers page is the only part of the whole system that candidates see, and it sits between every board posting and every application.

Three things about it decide whether the distribution reach translates into applications.

It has to look like you. A candidate who clicks from a job board arrives on a page that either continues your brand or announces that hiring has been outsourced to a system. The branded, mobile-friendly build is the point of the feature; the value is that somebody in marketing does not have to be involved for you to have one.

Mobile is not a nice-to-have. A large share of job board traffic is people looking on a phone during a break. If the application requires a desktop, a file upload from a laptop, or a fifteen-field form, you will lose a majority of the interest that your two hundred boards generated, and none of that loss will appear in any report — the applications simply never happen. In the trial, apply for your own role on your own phone, all the way through. Almost nobody does this, and it is the highest-value fifteen minutes of the whole evaluation.

The application form is the cheapest screening you own. Every custom question you add before a human reads anything is a filter you designed deliberately, in advance, rather than one you apply while tired at CV number ninety. Right to work, salary band, the certification the role legally requires, notice period. Three or four sharp questions remove a slice of the pile at zero cost and with a rule you can defend.

The tension is real: every question you add costs you some applications. That is fine when the question is genuinely disqualifying and expensive when it is curiosity. The test is whether a wrong answer would actually end the process. If it would not, take it out of the form and ask it in a screening call.

Implementation: what the first three weeks contain

No vendor's timeline survives contact with an organisation, but the work is predictable and it is worth budgeting because it lands on somebody who already has a job.

Week one is decisions, not configuration. Which pipeline stages you actually use, who sees what, which email goes out at which stage, and — the big one from the section above — whether this system or your existing one owns the employee record. Every hour spent settling these before touching the software saves three afterwards. The most common implementation failure in this category is configuring a workflow that nobody had agreed on.

Week two is data and connections. Careers page live, the boards that matter connected, the templates written, the integrations you rely on tested with real data rather than sample data. If you are bringing candidate history from a previous system, this is where you discover what the export actually contained, which is why question six on the sales list matters at both ends of the relationship.

Week three is one real role, run properly. Not a pilot in a sandbox. A live opening, with real applicants, read inside the product, with the people who will use it every week doing the using. Everything that is wrong with your configuration surfaces in that first role and almost none of it surfaces before.

Two costs that are not in anybody's quote. The first is training the hiring managers, who are the people most likely to bypass a new system and email you a CV directly — and the fix is not a training session, it is making the system the only route by which a candidate can be considered. The second is the month during which two systems are live, because you cannot switch off the old one until the roles running in it have closed. Plan for the overlap rather than being surprised by it.

Reporting, and which numbers are worth watching

The published reporting covers pre-built reports plus a custom report builder, spanning pipeline, time-to-hire and diversity metrics. That is the standard set for the tier, and the question is not whether the reports exist but which of them changes a decision.

Time-to-hire is the headline and the least actionable on its own. As a single number it tells you it takes too long and nothing about where. Broken into stage durations it becomes useful immediately, because delay in this process is almost always queueing rather than work — days waiting for an interviewer's calendar, days waiting for a decision that takes ten minutes once somebody sits down. Ask in the demo whether stage-level duration is available without building a custom report.

Source effectiveness is the one that pays for itself. Which boards produced candidates you actually interviewed, not which produced the most applications. Those two lists are usually very different, and the gap between them is where your premium board spend either justifies itself or does not. This is the report to run after your first quarter and act on before your next renewal.

Pipeline conversion between stages tells you where the process leaks. If ninety per cent of the people you phone-screen go to interview, your phone screen is not screening. If ten per cent do, your CV reading is not working. Neither is visible from a completion count.

Diversity reporting is a legal and a design question before it is a reporting one. The report can only show you what the process collected, at the points it collected it. Deciding what to collect, when, and who can see it is a conversation to have with somebody who knows the rules in your jurisdiction, not a switch to turn on because the feature exists.

Three things you will still do by hand

True of this product and of every product in the category. Worth listing so the business case does not assume otherwise.

Writing the criteria. No platform knows what good looks like for your role. Somebody has to decide the three or four things that actually matter and write them down before the pile arrives, and that decision is where hiring quality is really determined. Every ranking feature in every product is downstream of it.

Deciding. The shortlist is a suggestion. Choosing between the top three is judgement, and the platform's contribution ends before it. Any vendor implying otherwise is describing a product they do not have.

Keeping candidates warm. Templates and automated emails handle the mechanical part — acknowledgement, rejection, scheduling. The message that keeps a strong candidate engaged during a slow week is written by a person, and the strongest candidates are the ones who need it, because they are also in three other processes.

Where it sits against the other names on your shortlist

The shortlist is usually Workable, Greenhouse and Lever, and the honest way to separate them is not a feature grid, because at this level the feature lists converge. It is to ask what each product's centre of gravity is, and then check whether that centre matches your bottleneck.

Workable's centre is distribution and breadth. Post everywhere from one place, run the pipeline, and have the HR layer sitting underneath if you want it. That is the strongest fit for a company that hires across many roles and does not have a dedicated recruiting function, because breadth substitutes for specialisation.

The other two sit differently, and they are covered properly in their own pieces rather than summarised here in a way that would go stale — what Greenhouse does well and what it charges for and the honest read on Lever. The reason to read them is not to build a matrix. It is to work out which product's centre matches your actual constraint, because a product that is excellent at the thing you are not struggling with is an expensive way to keep struggling.

One category distinction that saves a lot of confusion at this stage: an applicant tracking system manages people who applied, and a candidate relationship system manages people who have not applied yet. Products in this tier increasingly do both, which is convenient and makes the comparison harder, because two vendors can both say yes to the same question while meaning different things. The difference between the two categories is the fastest way to stop talking past each other in a demo.

If you are earlier than a shortlist and still working out what the category contains at all, the plain overview of applicant tracking software is the piece that comes before this one.

Six questions for the sales call

Ask these, get the answers in writing, and note what each answer tells you beyond its content.

1. What is the fair usage limit on active jobs, and what happens when we cross it? Tests whether "unlimited" survives contact with a number. A vendor who knows their own threshold is a vendor whose contract you can read.
2. What is our price at the bracket above ours? You will grow. Get the next bracket's figure now, while you have leverage, rather than at renewal.
3. Which of the boards that matter to us are free and which are pass-through cost? Turns "200+ job boards" into a number that applies to your market.
4. Show me a scored candidate and tell me what the score was based on. The single most informative demo request in this whole category. Watch whether the answer points at the CV or at the model.
5. If we keep our existing HR system, what exactly do we switch off in yours? Forces the overlap question into the open before implementation rather than after.
6. What does the full candidate export contain, field by field? Not "can we export" — the field list, including application answers, scores, notes and email history. This is your exit, and the time to price it is before you enter.

Who it fits, and who should look elsewhere

It fits a company between roughly twenty and five hundred people, hiring across a range of roles, without a dedicated recruiting team, whose bottleneck is getting roles seen by enough people. It fits especially well if you have no HR system yet, because one purchase then covers two problems with no integration between them.

It fits less well if you are large and hire rarely, because you pay for your size rather than your usage. Or if you already run an HR system you are happy with, because you are buying a second one whether you use it or not. Or if your bottleneck is not distribution but reading — two hundred applicants per role and nobody with time to get through them, which more job boards makes worse rather than better.

That last case is the common one and the one people misdiagnose most often, because a full pipeline feels like success right up until somebody has to read it.

Where Orova Recruit fits, and where it does not

Stated plainly, because the argument of this article is that unclear scope is what costs people money.

Orova Recruit does not compete with Workable on the thing Workable is best at. It does not post to job boards — not a smaller number of them, none. It has no candidate database and does not source: it starts when a CV has arrived. It is not an HRIS, holds no employee records, runs no onboarding, no time off and no payroll. It has no cognitive or personality assessments and no one-way video interviews. If distribution or an HR system is what you need, this is not the product and no amount of feature comparison changes that.

What it covers is the stretch between the CVs arriving and the decision being made. Up to 500 resumes per role, each file up to 10MB, including photographs and scans. Every one scored from 0 to 100 against criteria you write yourself, and every score has to quote the line in the CV that earned it — which is the answer to question four on the list above, made structural rather than promised. Mark a criterion required and a resume that misses it is out however high the total.

Then the work around the reading. Interview questions drafted per candidate, aimed at whatever the CV left unclear. Notes taken by voice while you talk, in any of 56 languages, with nothing to install. Candidates lined up side by side with the gaps visible. Scheduling that refuses to double-book a room or an hour. Candidate emails sent from your own company mailbox so the candidate sees your name. And an offer letter draft that leaves salary and start date blank for a human to fill in.

It ranks and shows the evidence; choosing and passing is still your click. The boundary is the moment you decide, and everything after that — offer administration, onboarding, the employee record — belongs to whatever HR system you run. If that system is Workable, this sits in front of it. If it is something else, the same is true. That is the whole design.

One consequence of that narrowness is worth saying out loud, because it cuts against the way most of this category is sold: it is a second line on your invoice rather than a consolidation of two. The argument for it is that the reading stretch is where the hours actually go, and a tool aimed at exactly that stretch does not need to justify a platform migration to earn its place.

Common questions

How much does Workable cost?

As published in August 2026, the entry figures are $299 a month or $3,588 a year for Standard, $599 or $7,188 for Premier, and $719 or $8,628 for Enterprise. Pricing varies by company size bracket, so those are the shape rather than your quote. Check the current page before you build a business case on any of it.

Is Workable priced per job?

Not on the current published model. Active jobs are unlimited across all plans, subject to fair usage limits, and the price is set by company size bracket. A lot of comparison writing still describes the old per-slot model, which is why it is worth checking rather than inheriting.

Are assessments and video interviews included?

Not at the base tier. Texting, one-way video interviews and assessments are add-ons on Standard, listed at $89, $109 and $59 a month. Premier and Enterprise include all three plus referrals and a performance and engagement module.

Does Workable replace our HR system?

It can. The base plan includes an HRIS with onboarding, time-off management and payroll prep. Whether it should is a decision about which system owns the employee record, and the failure mode is having two systems that both think they do. Decide the direction explicitly and keep the flow one-way.

How many job boards does it post to?

Over two hundred free and premium boards from one posting action, according to the published figure. Premium boards still charge — the platform gives you one place to buy and manage them rather than free access to paid inventory.

Can we try it before buying?

There is a fifteen-day trial with the complete Standard feature set and no credit card required. Use it on one real role rather than on a tour of the interface. A real role exposes the parts of a workflow that a demo skips.

What if our problem is reading CVs, not finding candidates?

Then distribution features will not help and may make it worse by increasing volume. Look at what the platform does after the applications arrive: how candidates are scored, whether the score is traceable to evidence, and how much of the reading is genuinely removed rather than reorganised.

What do we lose if we leave?

Whatever is not in the export. Ask for the field list before you sign, including application question answers, scores, notes and email history. Every vendor answers "you can export your data"; the useful answer is which fields, in what format, and how far back.

What to do with this

Before the demo, answer one question honestly: is your bottleneck getting roles seen, or getting through what arrives?

If it is the first, this product's centre of gravity is aimed at you, and the tier question comes down to whether you will use texting, video screening and assessments. Do that arithmetic on the numbers above before the call, so the upgrade conversation is a decision rather than a reaction.

If it is the second, distribution is not your purchase. Spend the demo on the four minutes that matter: ask them to show you a scored candidate, ask what the score was based on, and see whether the answer points at a line in the CV or at a model. That is the whole evaluation for a reading problem, and it takes less time than the slide deck.

And whichever it is, book the trial before the demo rather than after it. Fifteen days with the full Standard feature set costs you nothing and answers questions a scripted walkthrough is designed not to raise: what the reading actually feels like at volume, whether the mobile application survives a real phone, how long a hiring manager takes to find the thing they need. A demo shows you the product working. A trial shows you your own process running inside it, which is the only thing you are really buying.

Either way, get the fair usage threshold and the next bracket's price in writing. Those two sentences are worth more than any feature comparison, including this one, because they are the two that will surprise you later.