General AI assistant
Hands-onClaude AI
A general-purpose AI assistant used in recruitment for writing, analysis and structured thinking, with a real free tier and plans that scale on usage rather than on features.
The short answer
- Best for
- Vacancy copy, candidate analysis and structured thinking, done by one recruiter
- Not ideal for
- Anyone who needs to know their usage ceiling in advance: the limits are not published as numbers
- Price
- Free, then USD 17 a month on an annual plan or USD 20 monthly; teams from USD 20 a seat
- Closest alternative
- ChatGPT for RecruitingBetter supplied with ready-made recruiting prompts and templates, at a comparable price
At a glance
- Recruitment fitWeak
- Ease of useStrong
- ImplementationStrong
- Price transparencyFair
- Value for moneyFair
- ScalabilityFair
Why these assessments
- Recruitment fitWeak
- A general assistant. There is no candidate record, no vacancy object, no pipeline and no ATS integration; everything recruitment-shaped is something you bring to it.
- Ease of useStrong
- From our own use it follows complex instructions more reliably than the alternatives we have tried, which is the difference between one prompt and five.
- ImplementationStrong
- For the buyer on the card, one recruiter, adoption is opening a tab and typing, and our own use bears that out. A team rollout is a different question: SSO, admin controls and a policy on what may be pasted in are not covered by this state.
- Price transparencyFair
- Plan prices and seat prices are published clearly, but usage is what you are buying and the limits are described in relative terms such as 5x and 20x rather than as numbers.
- Value for moneyFair
- USD 17 a month buys a tool we use for real work, but you cannot model what you get for it: the ceiling that decides whether the plan is enough is not published.
- ScalabilityFair
- Team plans run from 2 to 150 people, and above that the answer is Enterprise, where the seat carries a separate usage bill at API rates.
Our reading of the evidence on this page. No score, no formula.
Where it sits
- Discover
- Outreach
- Screen
- Interview
- Decision
The SourcrLab bottom line
A general assistant that earns its place in a recruitment stack through writing quality and instruction following rather than through any recruitment feature. The ladder is unusual in that it is almost entirely about usage: Free gets you the product, Pro at USD 17 a month on an annual plan adds the surrounding tools and more models, and Max from USD 100 buys nothing new except more room. Team plans start at USD 20 a seat with admin controls and a statement that content is not used for model training, which is the line that matters when the thing being pasted in is a candidate's CV. Disclosure: parts of SourcrLab itself were built with this product.
Best fit
The recruiter who writes a lot and wants it not to sound generated. Long-form vacancy copy, a summary of twenty CVs against a brief, a boolean string reasoned out rather than guessed, a difficult reply drafted twice. From our own use it holds a complicated instruction over a long piece of work better than the alternatives we have tried.
Think twice if
If you need to know in advance how much you can do in a month. The plans are sold on usage and the usage is described relatively, so the honest answer to whether Pro is enough for your week is that you find out by hitting the limit. And nothing here is recruitment software: no candidate data lives in it, and anything you paste into it is a data-protection decision you are making.
Hands-on experience
Pieter Henderyckx's take
Recruiter and founder of SourcrLab
Claude is what I reach for when I need structured, thoughtful output. Compared to ChatGPT, the writing feels cleaner and more nuanced, less "AI-sounding" if that makes sense. I actually used Claude to help build the SourcrLab website itself, which tells you something about how much I trust it for real work. For recruiting specifically, I find it great at writing longer-form content, analyzing candidate data, and helping with strategic thinking. It follows complex instructions better than most alternatives. The downside? It's less well-known, so there are fewer recruiting-specific prompts and templates floating around. But honestly, that might be an advantage: you're less likely to sound like every other recruiter using the same ChatGPT template.
Why choose Claude AI?
A free tier that is the product, not a demo
Chat on web, mobile and desktop, web search, memory, file creation, connectors and extended thinking are all on the free plan. What paying adds is room and surrounding tools, not the core capability, so you can judge the writing before spending anything.
It follows a long instruction
From our own use this is the practical difference: a brief with six constraints comes back with six constraints respected, rather than the first two. For recruitment writing, where tone, length, audience and a list of must-nots all matter at once, that is where the time is saved.
Team plans say content is not used for training
Alongside SSO, central billing and admin controls, the Team tier states that content is not used to train models. For a recruitment team pasting CVs and interview notes, that sentence is the one to put in front of whoever signs off on your data processing.
Less templated output, because fewer people template it
It is less widely used in recruitment than the obvious alternative, so there are fewer recruiting-specific prompt packs circulating. From our own use that cuts both ways, and on the output side it is an advantage: you are less likely to send the message every other recruiter is sending.
Main trade-offs
The limits you are buying are not published as numbers
Max is sold as 5x and 20x usage, which is a multiple of a base figure that is not stated. That makes the central question — is this plan enough for how I work — impossible to answer before subscribing, and it is the main reason the price transparency here is only fair.
Nothing about it is recruitment software
No candidate record, no vacancy, no pipeline, no ATS integration. Every piece of context arrives because you pasted it, and every output leaves because you copied it. That is fine for writing and analysis and useless as a system of record.
Pasting candidate data is a decision, not a habit
A CV in a chat window is personal data leaving your systems. The Team tier's no-training statement helps, and the Enterprise tier adds custom data retention, audit logs and a compliance API. On the individual plans, that governance is you.
Enterprise moves you onto a usage bill
It is described as seat price plus usage at API rates, from USD 20 a seat. That is a different budgeting model from a flat subscription, and the variable half depends on how heavily your team actually works, which is exactly the number nobody has before they start.
Team plans stop at 150 people
The published range is 2 to 150. A larger organisation is an Enterprise conversation whether or not it wants one, and the premium seat at USD 100 a month is a fivefold step from the standard seat before that conversation even starts.
Pricing and buying reality
Free, then USD 17 a month on an annual plan or USD 20 monthly; teams from USD 20 a seat
- Free: USD 0. Chat on web, mobile and desktop, code generation, web search, memory, file creation, connectors and extended thinking
- Pro: USD 17 a month with the annual subscription discount, billed as USD 200 up front, or USD 20 billed monthly. Adds Claude Code, Cowork, Design and Science, unlimited projects, Research, more models and a Microsoft 365 integration
- Max: from USD 100 a month, in two tiers described as 5x and 20x usage. Adds higher output limits, early access to features and priority access during peak traffic
- Team: 2 to 150 people. Standard seat USD 20 per seat per month billed annually, USD 25 monthly. Premium seat USD 100 per seat per month billed annually, USD 125 monthly. Adds enterprise search, SSO, central billing, admin controls, and a statement that content is not used for model training
- Enterprise: described as seat price plus usage at API rates, from USD 20 a seat. Adds SCIM, audit logs, a compliance API, custom data retention, IP allowlisting, a HIPAA option and a security product in beta
- Usage limits are described in relative terms rather than published as numbers on any tier
- Prices in USD
What we do not know
- The actual usage ceiling on any plan, in requests, words or hours
- What the two Max tiers cost individually
- What the premium Team seat carries that the standard seat does not
- What an Enterprise usage bill comes to in practice
Checked on the vendor's pricing page, August 2026. Claude pricing page, individual, team and enterprise plans
How it fits your stack
Works with
- Whatever you paste into it, plus connectors and a Microsoft 365 integration from Pro
- Your own documents, through file upload and file creation
Likely replaces
- The blank page at the start of a vacancy text
- Reading twenty CVs to find the five worth reading properly
May overlap with
- ChatGPT, which does the same job with a larger library of recruiting prompts around it
- The AI writing assistants already built into your ATS or outreach tool
Check before buying
- Whether your organisation allows candidate data in a general AI tool at all
- Whether the free tier already covers your week, since the paid step is mostly room
- Whether you need the Team tier's no-training statement in writing
- How many seats you expect, given the 150-person ceiling on Team
Claude AI vs alternatives
Products a buyer would genuinely put next to this one, with the reason you would pick each instead.
Choose ChatGPT when the ecosystem matters more than the writing: there are far more recruiting-specific prompts, templates and integrations built around it.
Choose Jasper when the job is marketing-shaped employer branding copy at volume, with brand voice controls, rather than open-ended analysis.
Used Claude AI?
Tell other recruiters what worked, what did not, and who it suits. Positive, mixed or critical — all useful.
Evidence
- Plan and seat prices on both billing cycles — checked against a primary source: Claude pricing page, individual, team and enterprise plans
- Free tier contents — checked against a primary source: Claude pricing page, individual, team and enterprise plans
- Team range of 2 to 150 people — checked against a primary source: Claude pricing page, individual, team and enterprise plans
- Team tier states content is not used for model training — vendor claim, not independently verified: Claude pricing page, individual, team and enterprise plans
- Enterprise billed as seat price plus usage at API rates — checked against a primary source: Claude pricing page, individual, team and enterprise plans
- Max sold as 5x and 20x usage — checked against a primary source: Claude pricing page, individual, team and enterprise plans
- Writing quality and instruction following in real work — first-hand use: Pieter uses Claude for recruitment writing and analysis, and for building SourcrLab
- Numeric usage limits on any plan — not enough evidence
- Individual prices of the two Max tiers — not enough evidence
- What the premium Team seat adds — not enough evidence
Last reviewed: August 2026. How SourcrLab verifies information
Work at Claude AI?
Email us to verify this profile or suggest a factual update.