Your hiring plan, answered in AI.
The first MCP connector built for headcount data. Ask Claude about open roles, slipped hires, recruiter capacity or offers over budget. Answers come from the same governed data as your headcount365 dashboard, and each person sees only what their role allows.
How are we tracking against the 2026 hiring plan?
Here's where the 2026 plan stands:
| Roles | |
|---|---|
| Total plan | 142 |
| Confirmed starts | 61 |
| Accepted offers (not yet started) | 18 |
| Actively recruiting | 39 |
| Not started yet | 24 |
| Behind schedule | 11 |
By today you expected 84 hires to be in place or with accepted offers, and you're at 79, so 5 short. Most of the 11 roles behind schedule are in Engineering (6) and Sales (3). Want me to list them?
Illustrative example. Sample company and data.
Read-only. The AI can look but never change anything.
Permission-aware. Each person sees only the roles and fields they can see in headcount365.
Your company only. Every answer is scoped to your company's data.
Every query logged. Admins can audit and revoke access at any time.
Your teams already use AI. Your headcount data shouldn't be the last thing they can ask it about.
Headcount is the largest line in most operating budgets and one of the most sensitive datasets a company holds. Today, getting it into an AI tool means exporting a spreadsheet and pasting it into a chat. The figures are stale as soon as you export them, and the file shows every column to whoever has it.
Spreadsheets leak.
An export ignores roles. Compensation, confidential requisitions and custom fields go to everyone who has the file, and to every AI chat it's pasted into.
Copies drift.
Once data is exported, nobody can say which version is right. Finance, TA and department heads each end up with a different count.
Questions wait.
"Are we on plan?" turns into a request to the one person who can pull the report, and the answer arrives days later.
headcount365 MCP gives AI a governed, live connection to your hiring plan. The data stays in headcount365, and the permissions come with it.
One connection. Your whole hiring plan.
headcount365 MCP connects AI assistants such as Claude to your hiring plan through the Model Context Protocol (MCP), the open standard AI tools use to connect to business systems. It brings in 85+ governed data points: plan, ATS and HRIS data side by side, plus your company's own custom fields.
Plan status in one question
Total plan, actively recruiting, accepted offers, confirmed starts and roles behind schedule. The counts use the same definitions as your headcount365 dashboard.
Find any requisition
Search and filter by stage, department, recruiter, plan year or start date, then open any role in full detail.
Plan vs. current vs. actual
Every role carries three versions: the originally approved plan, the current plan, and the actual figures from the accepted offer. Slippage and variance become something you can simply ask about.
From offer to employee
Once someone is hired, the role links to their HRIS record, so you can check start dates and pay against the plan.
Recruiter capacity
See each recruiter's open load against monthly capacity, adjusted for recruiters who are still ramping up.
Your org's open roles
A manager's direct reports and the open roles in their team, from your HRIS.
Built for everyone who owns a piece of the plan.
Talent Acquisition
Spend time on hiring, not on status reports.
- Answer "where are we on hiring?" yourself, without building a report.
- Spot recruiters who are over capacity before candidate experience suffers.
- See which roles are behind schedule and rebalance early.
- Give hiring managers answers without handing out spreadsheet access.
Finance & FP&A
Reconcile the plan in a single conversation.
- Compare approved plan, current plan and accepted offers role by role.
- Find offers above budget before they reach payroll.
- Track start-date slippage against forecast, since a hire starting later changes the year's spend.
- Get the same numbers as the dashboard every time, with no copy that has drifted.
Budget owners & department heads
Know your team's hiring without chasing anyone.
- "What's open in my org and when do they start?" — answered right away.
- See which of your roles are stuck and why.
- Ask follow-up questions in plain English, without learning another tool.
The business
Faster, safer headcount decisions.
- Executives get plan status on demand, consistent with the system of record.
- Sensitive data stays governed: the same permissions as the app, read-only, and audited.
- AI adoption goes forward without a new shadow-data problem.
Ask questions that used to take a report.
Plan status
How are we tracking against the 2026 hiring plan?
Here's where the 2026 plan stands:
| Roles | |
|---|---|
| Total plan | 142 |
| Confirmed starts | 61 |
| Accepted offers (not yet started) | 18 |
| Actively recruiting | 39 |
| Not started yet | 24 |
| Behind schedule | 11 |
By today you expected 84 hires to be in place or with accepted offers, and you're at 79, so 5 short. Most of the 11 roles behind schedule are in Engineering (6) and Sales (3). Want me to list them?
Illustrative example. Sample company and data.
Slippage (Finance)
Which Engineering hires have slipped the most from their original start date?
Five Engineering roles have moved more than 60 days from their originally planned start:
| Role | Originally planned | Now expected | Slip |
|---|---|---|---|
| Staff Platform Engineer | Mar 2 | Jul 13 | 133 days |
| Engineering Manager, Data | Apr 6 | Aug 3 | 119 days |
| Senior SRE | May 4 | Aug 17 | 105 days |
| ML Engineer | Jun 1 | Aug 31 | 91 days |
| Security Engineer | Jun 15 | Aug 24 | 70 days |
Three of the five are still in Approved and haven't been published yet, so recruiting hasn't started on them.
Illustrative example. Sample company and data.
Offers vs. budget (Finance)
Did any accepted offers this quarter come in above the budgeted salary?
Four accepted offers this quarter are above the current plan salary. I've compared each one in its own currency:
| Role | Currency | Plan | Offer | Over |
|---|---|---|---|---|
| Senior Account Executive | USD | 145,000 | 158,000 | +9.0% |
| Product Designer | GBP | 72,000 | 76,500 | +6.3% |
| Data Engineer | USD | 165,000 | 172,000 | +4.2% |
| Solutions Consultant | EUR | 88,000 | 90,000 | +2.3% |
The other 14 accepted offers this quarter are at or below plan.
Illustrative example. Sample company and data.
Recruiter capacity (TA)
Which recruiters are over capacity for the next three months?
Two recruiters are assigned more open roles than their capacity:
- Jordan Lee: 14 open roles against a capacity of 10/month. Most are Engineering.
- Sam Rivera: 9 open roles against 6/month. Sam is still ramping up, so capacity rises to 8 in December.
Alex Chen has room for about 5 more roles over the period. Moving the 3 Sales Development roles from Jordan to Alex would bring everyone within capacity.
Illustrative example. Sample company and data.
My org (budget owner)
What's open on my team and when are they expected to start?
You have 7 direct reports and 4 open roles in your org:
| Role | Stage | Expected start |
|---|---|---|
| Senior Product Manager | Published | Nov 17 |
| Product Analyst | Offer pending | Dec 1 |
| Associate PM | Approved | Jan 12 |
| Group PM, Platform | Pending approval | Feb 2 |
The Product Analyst offer is out now. The Group PM role still needs approval before recruiting can start.
Illustrative example. Sample company and data.
Offer to HRIS (TA + Finance)
For the Senior Account Executive hire, does the HRIS match the offer?
Mostly. Here's the comparison:
| Offer (ATS) | HRIS | |
|---|---|---|
| Title | Senior Account Executive | Senior Account Executive |
| Start date | Oct 6 | Oct 13 |
| Base pay | 158,000 USD / year | 158,000 USD / year |
The HRIS start date is one week later than the date in the accepted offer. It may be worth confirming which date is correct for payroll.
Illustrative example. Sample company and data.
Built for your most sensitive dataset.
Headcount data includes compensation, confidential roles and org structure. We designed the connector so that AI access is never broader than app access.
The AI sees exactly what the signed-in person sees in headcount365, down to the field. If compensation is hidden for your role, it's hidden in AI too.
Nothing in the connector can create, edit or delete data. Changes are made in headcount365.
People connect with their own headcount365 login and approve access on a consent screen. Nobody copies or shares API keys.
Every request is scoped to your company. Data never crosses between companies.
Every query is logged. Admins can switch access on for specific people or teams and revoke it at any time.
headcount365 doesn't cache or store your data outside your account to answer questions. Each answer is read live.
The data accuracy you expect from your most sensitive dataset, with the governance to match.
[PLACEHOLDER — e.g. "Teams already asking their hiring plan anything"]
"[PLACEHOLDER QUOTE]"
"[PLACEHOLDER QUOTE]"
[PLACEHOLDER — e.g. "X hours saved per month on plan reconciliation"]
Live in minutes.
- 01
Add the connector.
Paste one headcount365 URL into Claude (or another supported AI tool) as a custom connector.
- 02
Sign in and approve.
Log in with your headcount365 account and approve read-only access.
- 03
Ask.
Ask about your hiring plan in plain English. Answers respect your permissions automatically.
Frequently asked questions
What is MCP?
The Model Context Protocol is an open standard that lets AI assistants connect securely to business systems. headcount365 MCP uses it to give AI tools governed, read-only access to your hiring plan.
Which AI tools does it work with?
Claude (on the web, desktop and Cowork), Claude Code and Cursor. Other AI tools that support remote MCP connectors with standard sign-in should also work. Ask us about yours.
Can the AI change our data?
No. The connector is read-only. It can search and read your hiring plan, but nothing it does can create, edit or delete anything. All changes happen in headcount365.
Who can see what?
Exactly what each person can already see in headcount365. Requisitions, confidential roles, compensation and custom fields follow the same role-based permissions as the app. Two people asking the same question may get different answers, and that's intended.
What data can it answer questions about?
Requisitions and planned hires, hiring plan progress, recruiter workload, and managers' reporting lines. That includes the approved plan, the current plan, accepted offer details from your ATS, and the hired employee's record from your HRIS.
Do we need an ATS or HRIS connected?
Your hiring plan works on its own. Connecting your ATS adds the accepted offer details, and connecting your HRIS adds the hired employee details and reporting lines. headcount365 works with Greenhouse, Ashby, Gem and many other ATS and HRIS platforms.
Is our data used to train AI models?
headcount365 doesn't store your data in the AI tool or send it anywhere for training. How conversations are handled is set by your AI provider and your plan with them. Most business plans don't use customer conversations for training by default. [LEGAL REVIEW — confirm wording]
How accurate are the answers?
Answers are read live from headcount365, with no copy that can go stale. Plan status counts use the same definitions as the headcount365 dashboard. As with any AI assistant, check important figures, and you can always open the source record in headcount365.
Can we control who has access?
Yes. Admins switch access on for the whole company or for specific people and teams, can revoke it at any time, and every query is logged.
How long does setup take?
A few minutes per person. There's nothing to install and no API keys to manage.
What does it cost?
[PLACEHOLDER — e.g. "Included with headcount365 [plan name]. Book a demo for details."]
See your hiring plan answer back.
Book a 30-minute demo and we'll show headcount365 MCP working on a sample plan. Then bring your own questions.
Book a demo