# People Partner — Full Site Content > Concatenated full content of peoplepartner.io — homepage, pricing, FAQ, download details, and every blog post in one markdown document for AI ingestion. *Canonical URL: https://peoplepartner.io/llms-full.txt* *Last updated: 2026-05-02* ## What it is People Partner is the HR platform that runs on your Mac. Performance reviews, employee data, document Q&A, and AI guidance — and your data never leaves the device. ## Who it's for - Solo HR practitioners running the entire function - Founders managing HR alongside the rest of the company - Accidental HR — operations leads, EAs, finance folks who got handed the people work - Small-business operators with 10 to 200 employees ## What it does - **Performance management with AI-extracted insights** — pull review data into one place, surface themes across cycles - **Import your roster from any HRIS export** — CSV, Excel, PDF, Word, plain text - **Point it at your handbooks** — it reads them so you don't have to re-paste - **Built-in eNPS tracking** - **Early warning signals for team health** - **Representation insights with built-in privacy protection** - **Automatically redacts sensitive data before it touches AI** — SSNs, financial details, etc. - **Works with Claude, OpenAI, or Gemini — your choice** (BYOK) ## How it's different It is not a replacement for BambooHR, Gusto, or Rippling. It's the AI knowledge layer that sits on top of whatever HRIS you use. You export your data, import it into People Partner, and now you have one place where all your HR context meets AI. It is not a chatbot that gives generic answers. People Partner reads your actual data — the policies you wrote, the employees you hired, the reviews you ran — and gives specific, informed guidance. It is not a cloud SaaS. Your database lives on your Mac. Sensitive data is redacted before any AI provider sees it. ## Pricing $99 one-time purchase. BYOK (bring your own API key). Typical AI usage runs $2 to $8 per month, paid directly to your provider. No subscriptions, no per-seat fees, no enterprise upsells. ## Requirements - macOS 12 (Monterey) or later - Apple Silicon or Intel - An API key from OpenAI, Anthropic, or Google (about 2 minutes to set up) ## Links - [Pricing](https://peoplepartner.io/pricing.md) - [FAQ](https://peoplepartner.io/faq.md) - [Download](https://peoplepartner.io/download.md) - [Blog](https://peoplepartner.io/blog) --- # Pricing ## Price **$99 one-time purchase.** That's it. No subscription, no per-seat, no enterprise upsell. ## What's included - The People Partner desktop app for macOS (Intel and Apple Silicon) - All current features: performance reviews, employee data, eNPS, document Q&A, PII redaction, AI guidance - Free updates for the lifetime of the product - Works with Claude, OpenAI, or Gemini — pick your provider, swap any time - Email support at hello@peoplepartner.io ## What you also need You bring your own API key from OpenAI, Anthropic, or Google. Typical usage runs **$2 to $8 per month** depending on how often you query and which model you pick. You pay your provider directly — we never see the bill, and we never see your data. Setup takes about 2 minutes. A step-by-step guide ships with the app. ## What you don't pay for - No monthly subscription - No per-seat licensing - No "AI add-on" upcharge - No vendor lock-in — switch providers any time - No data resale (we don't have your data) ## Buy Visit https://peoplepartner.io/upgrade to purchase. License is delivered by email and activates the app on your Mac. --- # Download ## Download People Partner People Partner is a desktop app for macOS. Download the .dmg, double-click to install, paste your license key from your purchase email. ## System requirements - macOS 12 (Monterey) or later - Apple Silicon or Intel Mac - About 200 MB free disk space - Internet connection only required when querying AI providers (everything else works offline) ## Install 1. Visit https://peoplepartner.io/download 2. Download the .dmg 3. Open it and drag People Partner to your Applications folder 4. Launch the app 5. Enter your license key from the purchase email 6. Add an API key from OpenAI, Anthropic, or Google (about 2 minutes) 7. Import your first roster from any HRIS export ## Trial A free trial is available before purchase. The trial runs through a Cloudflare Worker proxy with a 50-message cap so you can test the AI features without setting up a key. ## Apple notarization People Partner is signed and notarized by Apple. macOS Gatekeeper accepts it without warnings. The app is built and signed via GitHub Actions; the developer certificate is registered to People Partner's parent company, FoundryHR LLC (Apple Team ID NNC7N645Q8). --- ## Frequently Asked Questions ### How much does it cost? $99 one-time. That’s it. Bring your own AI API key — we support OpenAI, Anthropic, and Google (typically $2-8/month usage). No subscriptions, no per-seat fees, no enterprise upsells. ### Do I need my own API key? Yes. Bring your own API key from OpenAI, Anthropic, or Google — whichever you prefer. Setup takes about two minutes and typical usage costs $2-8/month. A step-by-step guide is included. ### What data can I import? Employee rosters, performance review exports, eNPS survey results, and HR documents like handbooks and policies. Supports CSV, Excel, PDF, Word, and plain text. ### What happens if I accidentally include an SSN? It’s automatically redacted. The app scans every message for SSNs, credit card numbers, and bank account info. If found, it replaces them with placeholders like [SSN_REDACTED] and shows a brief notification. ### Can I use this alongside my existing HR tools? That’s exactly how it’s designed to work. Export data from your review tool, survey platform, or HRIS and import it into People Partner. It’s the layer where all your HR data meets AI. ### Does People Partner work offline? Yes. You can browse and manage your employee data, performance reviews, and documents without an internet connection. AI features require a connection to your chosen provider (OpenAI, Anthropic, or Google), but everything else works offline. ### Is my employee data sent to the cloud? Your database lives entirely on your Mac. When you ask the AI a question, the relevant context is sent to your chosen AI provider — but sensitive data like SSNs and financial details are automatically redacted before anything leaves your machine. ### How does the AI work with my HR data? When you ask a question, People Partner finds the relevant employee records, review data, or policy documents and includes them as context for the AI. The AI reads your actual data to give specific, informed guidance — not generic HR advice. ### What Mac versions are supported? People Partner requires macOS 12 (Monterey) or later. It runs natively on both Intel and Apple Silicon Macs. ### What file formats can I import? CSV, Excel (.xlsx), PDF, Word (.docx), and plain text files. The smart import pipeline includes HRIS presets, intelligent column mapping, and duplicate detection to make setup fast. ### How is People Partner different from BambooHR or Gusto? BambooHR and Gusto are cloud HRIS platforms for payroll, benefits, and onboarding. People Partner is a local AI knowledge layer — it pulls data from those tools (and spreadsheets, documents, surveys) into one place and pairs it with an AI that understands your team. Most users run People Partner alongside their existing HRIS, not instead of it. ### Can I use People Partner alongside my existing HRIS? That’s exactly how it’s designed. Export data from your HRIS, review tool, or survey platform and import it into People Partner. It becomes the layer where all your HR context meets AI — complementing, not replacing, your existing tools. ### Is there a subscription or recurring fee? No. People Partner is a one-time $99 purchase. You own it forever. The only ongoing cost is your AI API usage (typically $2-8/month), which you pay directly to your chosen provider. --- # Blog # How People Partner Manages Context > A technical deep-dive into the 7-stage pipeline that turns every question into an accurate, grounded HR answer — from PII scanning to answer verification. *Canonical URL: https://peoplepartner.io/blog/how-people-partner-manages-context* *Published: 2026-03-03* *Keywords: ai context pipeline, hr ai architecture, local ai processing, token management, pii scanning, ai answer verification* ## The Pipeline (Every Message) Every message you send goes through 7 stages before you see a response: **User Input → PII Scan → Query Classification → Context Assembly → Token Budget → LLM → Answer Verification → Response** All of it orchestrated between the React frontend and the Rust backend. No cloud preprocessing. No external API calls until the LLM itself. Here's how each stage works. --- ## Stage 1: PII Scanning Before anything touches the AI, the user's input is scanned for financial PII — Social Security numbers, credit card numbers, bank account patterns — using regex in Rust. If detected, the text is auto-redacted and replaced with placeholders. The user sees a notification, never a blocking modal. Security without friction. `"John's SSN is 123-45-6789"` → `"John's SSN is [SSN REDACTED]"` The redacted version is what gets sent to the LLM. The original is never logged. ## Stage 2: Query Classification The system classifies every query into one of 6 types using keyword heuristics — no LLM call needed: | Query Type | Example | What It Retrieves | |---|---|---| | Individual | "Tell me about Sarah" | Full employee profiles (max 3) | | Comparison | "Top performers in Engineering" | Multiple full profiles (max 8) | | List | "Who's in Marketing?" | Lightweight summaries (max 30) | | Aggregate | "What's our headcount?" | Org-wide stats only, no individual data | | Attrition | "Who left this year?" | Recent terminations with full context | | General | "How should I handle this PIP?" | Balanced sample (max 5) | The classification uses priority-based logic: explicit names always win (→ Individual), ranking keywords beat department keywords (→ Comparison over List), and so on. This means the system retrieves exactly the right depth of data for every question — no wasted tokens, no missing context. ## Stage 3: Query Mention Extraction A heuristic NER-like system extracts structured signals from the raw query: - **Employee names** — Capitalized words that aren't common English/HR terms, with possessive handling ("Sarah's" → "Sarah") - **Department names** — Word-boundary matching ("IT" matches only at word boundaries, not inside "wITh") - **Intent flags** — `is_performance_query`, `is_tenure_query`, `is_top_performer_query`, `is_theme_query`, and more - **Tenure direction** — Distinguishes "who's been here longest" vs "newest hires" vs "upcoming anniversaries" - **Theme detection** — Maps semantic variants to canonical themes ("people skills" → communication, "coaching" → mentoring) - **Target field** — Whether theme queries target strengths ("excels at") or opportunities ("struggles with") All of this happens in Rust, in microseconds, with zero network calls. ## Stage 4: Context Assembly Based on the query type, the system assembles context from 5 sources: ### 4a. Company Context Single SQL query: company name, state, industry, active employee count, department count. ### 4b. Organization Aggregates A battery of SQL queries computed for every message (~2K characters when formatted): - Headcount breakdown (total, active, terminated, on leave) - Department distribution with percentages - Performance rating distribution (exceptional / exceeds / meets / needs improvement) - eNPS score with promoter/passive/detractor breakdown - YTD attrition stats (voluntary/involuntary, average tenure, annualized turnover rate) This is what makes aggregate queries accurate — the LLM doesn't guess, it reads pre-computed ground truth. ### 4c. Employee Context (Query-Type-Adaptive) This is the key innovation. Different queries retrieve different depths: - **Individual:** Full profile with all ratings, all eNPS scores, career summary, extracted review highlights (strengths, opportunities, themes, quotes), and trend analysis - **Comparison:** Same full profiles but for multiple employees, with specialized retrieval for top/bottom performers or theme-based filtering - **List:** Lightweight summary structs (~70 characters each) — just name, department, title, status - **Aggregate:** No individual employee data at all — the org aggregates are sufficient - **Attrition:** Recent terminations with termination reason, tenure, and performance history Employee retrieval priority: selected employee (from UI panel) → name matches → department matches → specialized queries → random sample fallback. ### 4d. Cross-Conversation Memory Hybrid search for up to 3 relevant past conversation summaries: 1. First tries summary-only FTS5 search (more focused) 2. Falls back to full conversation FTS if no summary matches Summaries are 2–3 sentence AI-generated distillations of past conversations. No vector database. No embedding API calls. SQLite FTS5 — works completely offline. ### 4e. Document Context FTS5 search across ingested company documents (employee handbook, policies, etc.). Returns ranked chunks with source metadata for citation. Token-budgeted to avoid crowding out other context. ## Stage 5: Token Budget Management The system operates within a 200K context window: | Allocation | Tokens | Notes | |---|---|---| | System prompt | 20K | Persona + company + employees + docs + memory | | Conversation history | 150K | 75% of window | | Output reserved | 8K | Space for the response | | Safety buffer | 22K | Headroom | Within the system prompt budget, each query type gets a different token allocation: | Query Type | Employee Tokens | Memory Tokens | Total | |---|---|---|---| | Aggregate | 0 | 500 | 1,000 | | Individual | 4,000 | 1,000 | 5,000 | | Comparison | 3,000 | 500 | 3,500 | | List | 2,000 | 500 | 2,500 | | Attrition | 2,000 | 500 | 2,500 | | General | 2,000 | 1,000 | 3,000 | When conversation history exceeds the budget, the oldest user/assistant message pairs are silently dropped. This preserves recent context without notification — a deliberate UX decision. The user shouldn't have to worry about context management. Employee context is capped at 10 employees and 16K characters (~4K tokens) to prevent any single query from blowing the budget. ## Stage 6: System Prompt Assembly Everything gets composed into a structured system prompt: - **Persona preamble** — The selected AI persona with company-specific variables - **Communication style** — Persona-specific guidelines - **Company context** — Name, state, employee count, departments - **Organization data** — Formatted aggregates (headcount, ratings, eNPS, attrition) - **Context awareness rules** — State-specific employment law guidance, employee name references, document citation preferences - **Boundaries** — Guidance not legal advice, recommend counsel for litigation - **Relevant employees** — Formatted profiles or summaries - **Relevant documents** — Document chunks with source citations - **Relevant past conversations** — Memory summaries ## Stage 7: Answer Verification Post-response, not pre-response. For aggregate queries only, the system: 1. Extracts numeric claims from the AI's response using regex (headcount patterns, percentages, ratings, eNPS scores) 2. Compares each claim against the pre-computed ground truth aggregates 3. Returns a verification result: **Verified**, **Partial Match**, or **Unverified** 4. The frontend displays a verification badge on the message This catches hallucinated numbers — if the AI says "82 active employees" but the database has 85, it's flagged. No silent errors. ## Background Operations These run on a default model regardless of user selection, keeping costs predictable: | Operation | Trigger | Model | |---|---|---| | Conversation summaries | After conversation ends | Default (Sonnet) | | Title generation | After first message | Default (Sonnet) | | Highlight extraction | After review import | Default (Sonnet) | | Employee career summaries | After highlights exist | Default (Sonnet) | The model selector only affects interactive chat. Background operations always use the cost-efficient default. ## Model-Aware Adaptation When users switch models (e.g., Sonnet → Opus), the system adapts automatically. The selected model's context window determines the conversation token budget (75% of window). Gemini's 1M context window gets 750K for conversation; OpenAI's 128K gets 96K. Max output tokens are passed through to the provider. ## What Makes This Different 1. **Zero-LLM-call context routing** — Query classification and employee retrieval happen entirely in Rust with SQL and heuristics. No "pre-processing" LLM call. 2. **Query-adaptive retrieval** — An aggregate question doesn't waste tokens on individual profiles. A name query doesn't waste tokens on org-wide stats. 3. **Ground truth verification** — The LLM's numeric claims are checked against SQL-computed facts post-response. 4. **PII never leaves the device unredacted** — Scanning happens before context building, in Rust, with no network calls. 5. **Cross-conversation memory without embeddings** — SQLite FTS5 hybrid search. No vector database, no embedding API calls. Works offline. 6. **Token observability** — Every query tracks retrieval metrics: employees found/included, memories found/included, token budget vs actual usage, retrieval time in milliseconds. The result: every answer is grounded in your actual data, verified against your actual numbers, and processed entirely on your machine. --- [People Partner](https://peoplepartner.io) is a local-first HR knowledge platform that runs entirely on your Mac. Your employee data, policies, and documents stay on your machine — processed by on-device AI, never uploaded to the cloud. $99 one-time purchase. --- # How to Run Performance Reviews at a Small Company (Without Enterprise Software) > A practical guide to performance reviews for startups and small businesses. Skip the enterprise tools — use a lightweight framework and AI to spot patterns. *Canonical URL: https://peoplepartner.io/blog/how-to-run-performance-reviews-small-company* *Published: 2026-03-03* *Keywords: performance reviews small company, how to do performance reviews startup, small business performance review, lightweight performance review* ## Why Nobody Does Reviews (And Why That's a Problem) At most small companies, performance reviews don't happen. Not because anyone decided they shouldn't — but because nobody set them up, nobody owns the process, and there's always something more urgent. It feels like bureaucracy. The kind of thing big companies do because they have HR departments with nothing better to do. When you're a 15-person startup, spending a week on formal reviews feels like a luxury you can't afford. So you skip them. Then you skip them again. Then a year goes by and you're sitting across from an employee who's genuinely surprised they're being let go — because nobody ever told them there was a problem. Or a top performer leaves for a competitor because they never heard they were valued, never got a raise conversation, never felt like anyone noticed their work. The cost of not doing reviews isn't visible until it's expensive. ## Why Small Companies Skip Reviews (And Shouldn't) Let's be honest about the reasons. And then let's be honest about why they don't hold up. **"We're too small for that."** You're too small for a formal, enterprise-style review process. You're not too small for telling people how they're doing. Those are different things. **"We give feedback all the time."** Maybe. But informal feedback doesn't create documentation. When you need to make a termination decision, "I've told them multiple times" without any written record is a legal liability. **"The tools are too expensive."** Lattice starts at $11/person/month. Culture Amp is similar. For a 30-person company, that's $4,000+ per year for performance management software. That's a real expense. But you don't need those tools. You need a process. **"Nobody knows how to do it."** This one's fair. Most founders and small-company managers have never been trained on giving reviews. The good news: the bar is low, and a simple framework is better than no framework. Here's what happens when you don't do reviews: - **Surprise terminations.** If the first time someone hears about a performance problem is when they're being fired, you have a problem — and potentially a lawsuit. - **No documentation trail.** When a terminated employee files a complaint, your defense is "we talked about it." Without documentation, that's not a defense. - **Good people leave quietly.** Your best performers want to know they're valued. Silence isn't neutral — it feels like indifference. They'll leave for somewhere that notices them. - **Problems fester.** A small issue in Q1 becomes an entrenched behavior by Q4 because nobody addressed it. The conversation gets harder every month you wait. ## A Lightweight Framework You don't need a 10-page review template. You need three questions, 30 minutes, and a place to write it down. ### The Three Questions Every review — whether it's quarterly, biannual, or annual — can start with these: 1. **What went well this period?** Be specific. Name projects, behaviors, outcomes. "You did a good job" is worthless. "You led the migration project under deadline and kept the team aligned through two scope changes" is useful. 2. **What could improve?** Again, specific. Not "be more proactive" — that means nothing. "When you encounter a blocker, I'd like you to flag it in standup rather than waiting until the weekly check-in." Actionable, observable, fair. 3. **What do you need from me?** This is the question most managers skip, and it's the most important. It shifts the review from a one-way judgment to a two-way conversation. You'll learn things you wouldn't otherwise: that they need clearer priorities, that they're struggling with a coworker, that they want to grow into a different role. ### The Format Keep it simple. A shared Google Doc works. A Notion page works. A plain text file works. The format doesn't matter. What matters: - The manager writes their notes before the meeting - The employee gets a chance to self-reflect (send them the three questions 48 hours in advance) - The conversation happens live — don't just send a document - Both parties can see the final written summary - It's stored somewhere you can find it later ### The Cadence Quarterly is ideal for small companies. It's frequent enough that nothing festers, and short enough that each review doesn't feel like a massive undertaking. If quarterly feels like too much, start with every six months. Annual reviews are better than nothing, but a lot can go wrong in 12 months of silence. Block 30 minutes per employee. That's it. For a 20-person company with quarterly reviews, that's 10 hours per quarter — roughly one day spread across a couple of weeks. That's manageable. ## How to Track Without Enterprise Software You don't need Lattice. You don't need Culture Amp. You don't need 15Five or Betterworks or any other performance management platform. Here's what actually works at small scale: **Under 10 employees:** A Google Doc per employee. One running document with dated entries. The manager adds notes after each review. Search works well enough at this scale. **10-25 employees:** A shared folder with a standard template. Each review is a new entry in the employee's folder. Consider a simple spreadsheet that tracks: employee name, last review date, next review date, key themes. **25-50 employees:** This is where manual tracking starts to break down. You're spending more time managing the process than having the conversations. At this point, you need either a dedicated tool or a smarter system. The critical requirement at every scale: **consistency**. Use the same template. Follow the same cadence. Store reviews in the same place. The system fails when every manager does it differently and files end up scattered across Google Drive, email, and Slack messages. ## Using AI to Spot Patterns This is where things get interesting — and where you can get leverage that even expensive enterprise tools struggle to provide. Once you have a few cycles of reviews documented, patterns start to emerge. But you won't see them by reading individual reviews one at a time. You need something that can look across all of them at once. Import your review documents into an AI tool like People Partner. Then start asking questions: **"Who hasn't had a review in over six months?"** Simple accountability check that's surprisingly hard to answer from a folder of documents. **"What themes come up across the engineering team's reviews?"** Maybe three different engineers are all flagged for communication issues. That's not three individual problems — that's a team dynamic or a management issue. **"What feedback did we give this person in their last two reviews?"** Before you sit down for a review, you need to know what was said before. Continuity matters. If you flagged something in Q2, you should be following up in Q3. **"Help me prepare for a difficult conversation about performance."** Walk through the scenario with the AI. It knows the employee's review history, your company's policies, and the context. It can help you frame feedback constructively, anticipate pushback, and make sure you're covering the legally important documentation points. This isn't about replacing your judgment. It's about walking into every review prepared — with full context, clear patterns, and a plan. The managers who do this well spend 15 minutes preparing. Most managers spend zero. ## Getting Started This Week You don't need to build a complete system before you start. Here's a practical five-step checklist: **1. Pick a cadence.** Quarterly is recommended. Put the next review cycle on the calendar right now. Block the time. If it's not scheduled, it won't happen. **2. Write your three questions.** Use the ones above or adapt them. Keep it to three or four questions maximum. More than that and the conversation loses focus. **3. Create a template.** One page. Employee name, date, manager name, the three questions with space for notes, and a section for action items and follow-ups. Save it somewhere everyone can access. **4. Block 30 minutes per employee.** Send the calendar invites. Give employees the questions 48 hours in advance so they can reflect. Frame it as a conversation, not an evaluation. **5. Actually do it.** The first round is the hardest. It will feel awkward. The conversations might be stilted. Some employees will be nervous. That's normal. It gets better every cycle. The important thing is to start. ## The Bar Is Low. Clear It. Most small companies do nothing. No reviews. No documented feedback. No structured conversations about performance. The employees who need guidance don't get it. The employees who deserve recognition don't hear it. And when things go wrong, there's no paper trail. Even a basic, imperfect review process puts you ahead of the majority. You don't need enterprise software. You don't need a certification in performance management. You need three questions, 30 minutes, and the willingness to have honest conversations. Start this quarter. Your future self — dealing with a difficult termination, trying to retain a key employee, or defending a decision — will thank you. --- [People Partner](https://peoplepartner.io/download) helps you manage the entire review cycle — store past reviews, spot patterns across your team, and prepare for difficult conversations with AI that understands your specific context. Runs locally on your Mac. $99 one-time, no subscription. --- # I Already Use BambooHR. Do I Still Need People Partner? > People Partner isn't replacing your HRIS — it sits on top of BambooHR, Gusto, or Rippling as a knowledge and AI layer. Here's how they work together. *Canonical URL: https://peoplepartner.io/blog/people-partner-with-bamboohr-gusto* *Published: 2026-03-01* *Keywords: BambooHR alternative, use with BambooHR, HR knowledge layer, Gusto HR tool, HRIS complement* ## You Already Have an HRIS. That's Not the Problem. If you're using BambooHR, Gusto, or Rippling, you've already solved the operational side of HR. Payroll runs on time. Benefits are administered. PTO requests flow through an approval chain. Employee records have a home. That's real. That matters. And nobody's suggesting you rip it out. But there's a category of HR work your HRIS doesn't touch — and if you're honest with yourself, you've felt the gap. It shows up when a manager asks you how to handle a performance issue with someone who's been underperforming for six months, and you need to pull context from their last three reviews, their job description, your PIP template, and the relevant state employment law. Your HRIS has some of that data. But it doesn't connect it. And it definitely doesn't help you think through it. The question isn't "should I switch from BambooHR?" It's "is there a layer missing on top of it?" ## The Problem: Data Silos Within Your Own HR Stack Your HRIS is excellent at structured, transactional data. It knows that Sarah started on March 15, 2024. It knows her salary is $85,000. It knows she has 12 PTO days remaining. What it doesn't know: - That her last two performance reviews flagged communication issues with the engineering team - That your company handbook has a specific progressive discipline process she hasn't been placed on yet - That California (where she works remotely) has different final paycheck timing requirements than your headquarters state - That the manager asking you about her situation gave her a glowing review 18 months ago and might need coaching on consistency This information exists. It's in review exports, in policy documents, in the notes you took during a 1:1 with her manager. But it lives in five different places, and your HRIS isn't designed to pull it together. HRIS platforms are systems of record. They store data. They don't reason across it. ## What People Partner Adds on Top People Partner is a knowledge layer, not a replacement layer. Think of it as the difference between a filing cabinet and an analyst. Your HRIS is the filing cabinet — it stores everything in its proper place. People Partner is the analyst who can pull from every drawer at once and help you see the full picture. Here's what that looks like in practice: **Cross-referencing employee context.** Import employee data, past reviews, and your policies into People Partner. When you need to make a decision about someone, the AI can synthesize information across all of those sources. Instead of opening four tabs and manually piecing together a timeline, you ask a question and get an answer that accounts for the full picture. **Policy interpretation with specifics.** Your handbook says one thing. The employee's situation has nuances. People Partner can help you apply your actual policies to the actual scenario — not generic advice from the internet, but guidance rooted in your documents. **Pattern recognition across your team.** When you've imported review data for 50 employees, you can start asking questions like: "Which teams have the most inconsistent review scores between managers?" or "Who hasn't had documented feedback in over six months?" Your HRIS might have the raw data, but it's not built to surface these patterns. **Coaching and conversation prep.** Before a difficult conversation — a PIP, a termination, a compensation discussion — you can walk through the scenario with the AI. It knows the employee's history, your policies, and the relevant considerations. It's the preparation step most HR managers skip because it takes too long to do manually. ## How They Work Together The workflow is straightforward, and it doesn't require any integration or API connection. **Step 1: Export from your HRIS.** BambooHR, Gusto, and Rippling all support CSV or Excel exports of employee data, reviews, and other records. Run the export. **Step 2: Import into People Partner.** Drop the files in. Add your handbook, your policy documents, your review templates, any other HR documents you reference regularly. **Step 3: Your HRIS stays your system of record.** Payroll, benefits, PTO, onboarding workflows — all of that continues to live in BambooHR or Gusto or wherever it belongs. Nothing changes. **Step 4: People Partner becomes your thinking partner.** When you need to make a decision, answer a complex question, or prepare for a sensitive conversation, you go to People Partner. It's where you do the analytical and advisory work that your HRIS was never designed for. You update People Partner periodically — after review cycles, when policies change, when you onboard new employees. It's not a real-time sync. It doesn't need to be. The knowledge layer doesn't need to update every time someone requests PTO. It needs to be current on the things that matter for decision-making. ## Why Local Matters Here Here's the part most people don't think about until it's too late. To make People Partner useful, you're exporting employee data from your HRIS. Names, roles, compensation, performance reviews, disciplinary notes. This is sensitive information. With People Partner, that exported data stays on your Mac. It's processed by on-device AI. It never touches a cloud server. It never gets sent to a third-party API. Your HRIS vendor doesn't know you're using it. No additional data processor to add to your privacy policy. Compare that to cloud-based analytics tools that want you to connect your HRIS via API and pipe employee data through their servers. You're adding another vendor with access to your most sensitive information, another entry in your data processing agreement, another potential breach surface. Local processing means the gap between your HRIS and your knowledge layer doesn't create new privacy exposure. The data goes from one secure system (your HRIS) to another secure system (your machine). That's it. ## Who This Is For This isn't for everyone. If you're an HR team of 10 people at a 2,000-person company with Workday and a full tech stack, you probably have enterprise analytics tools already. This is for: - **Solo HR managers handling 50-300 employees.** You're the entire HR department. You use BambooHR or Gusto for the operational stuff, but the strategic and advisory work falls entirely on you. You need a thinking partner, not another admin tool. - **Founders doing HR without an HR team.** You chose Gusto for payroll because it was easy. But now you're dealing with performance management, policy questions, and employee relations issues that Gusto wasn't built for. You need help with the hard questions. - **Operations leads who inherited HR.** You're the accidental HR manager. Your HRIS handles the basics, but every complex situation sends you down a research rabbit hole. You need faster, more reliable answers that account for your specific context. - **Anyone who uses their HRIS for admin but needs help with judgment calls.** The HRIS handles the "what." You need help with the "what should I do about it?" ## Your HRIS Is Doing Its Job. This Fills the Gap It Wasn't Designed For. People Partner doesn't compete with BambooHR. It doesn't replicate Gusto's payroll. It doesn't try to replace Rippling's workflow automation. It does the thing those platforms can't: it takes your scattered HR knowledge — policies, employee history, reviews, documents — and turns it into something you can actually reason with. It's the layer between "I have the data" and "I know what to do." $99 one-time. Runs on your Mac. Your data stays on your machine. And your HRIS keeps doing exactly what it's good at. [Get People Partner](/upgrade) and make your existing HR stack smarter. --- # Why Your HR Data Shouldn't Live in the Cloud > The default for software is cloud-first. But employee data — SSNs, salaries, performance reviews — deserves a different conversation. *Canonical URL: https://peoplepartner.io/blog/hr-data-shouldnt-live-in-cloud* *Published: 2026-02-27* *Keywords: hr data privacy, employee data privacy, local hr software* ## The Default We Stopped Questioning At some point over the last decade, we collectively decided that all software should live in the cloud. And for most applications, that makes sense. Your project management tool, your CRM, your email — cloud works. But somewhere along the way, we applied the same default to employee data. Social Security numbers. Salary information. Performance reviews. Medical accommodation requests. Disciplinary records. Termination documentation. We put all of it on someone else's servers and didn't think twice. Maybe we should think twice. ## What "Cloud" Actually Means for Your Employee Data When your HR data lives in a cloud-based tool, here's what's technically happening: Your employees' most sensitive information is stored on servers owned and operated by a third party. That third party's employees — engineers, database administrators, support staff — have some level of access to those servers. Your data sits alongside data from thousands of other companies in a multi-tenant architecture. The cloud provider is responsible for: - Physical security of their data centers - Network security and encryption - Access controls for their own employees - Backup and disaster recovery - Compliance with whatever regulations apply You're responsible for: - Trusting that they do all of the above correctly - Managing who at your company has access - Not much else That trust isn't always warranted. ## The Breach Problem Let's talk about what keeps happening. HR and payroll platforms are high-value targets for attackers. They contain exactly the data that enables identity theft: names, SSNs, dates of birth, bank account numbers, addresses. Recent years have seen breaches at major HR-adjacent platforms affecting millions of records. These weren't small, careless companies — they were established vendors with security teams and compliance certifications. **The uncomfortable math:** When you use a cloud HR tool, your exposure isn't limited to your own security practices. You're exposed to the vendor's security, their infrastructure provider's security, and every other link in the chain. Your attack surface is their attack surface. A 15-person company using a cloud HRIS has the same vendor breach exposure as a 15,000-person company using the same platform. But the small company has far fewer resources to respond to a breach notification. ## The AI Complication This problem has gotten significantly worse in the last few years, and it's because of AI. Managers and HR leads are pasting employee information into ChatGPT and other AI tools to get help writing performance reviews, drafting PIPs, analyzing compensation data, and generating HR policies. Think about what's happening: someone copies an employee's performance history — including their name, role, specific behavioral issues, and salary — into a cloud AI model. That data is now part of a request sent to a third-party API, processed on third-party infrastructure, and potentially used for model training. This isn't hypothetical. It's happening right now, often without any formal policy addressing it. The employee's data touches at least two third-party systems, and neither asked for consent to process it through AI. ## The "We're Compliant" Argument Cloud HR vendors will tell you they're SOC 2 compliant, GDPR ready, and encrypted at rest and in transit. And they probably are. But compliance is a floor, not a ceiling. It means you follow a set of practices — not that you can't be breached, not that a rogue employee can't access data. Compliance certifications are about process, not guarantees. They're not a substitute for asking: does this data need to be on someone else's servers at all? ## What "Local-First" Means Local-first software keeps your data on your own hardware. Not a server in Virginia. Not a data center in Dublin. Your machine. Your network. Your control. For HR data, this means: - **No third-party access.** Nobody at a vendor company can see your employee records, even theoretically. The data doesn't exist on their infrastructure. - **No breach exposure beyond your own security.** If a cloud HR vendor gets breached, your data isn't affected because it was never there. - **No vendor lock-in on your data.** Your files are on your machine, in formats you control. If you stop using the software, the data is still yours. - **No ambiguity about data processing.** If the software runs locally, your data isn't being sent anywhere. Period. This isn't a new concept. It's how software worked for decades before the cloud era. We gave it up for convenience, collaboration, and ubiquitous access. Those are real benefits — but they come with real tradeoffs, and for employee data, the tradeoffs deserve scrutiny. ## This Isn't Anti-Cloud This isn't a blanket argument against cloud software. Cloud infrastructure powers most of modern business and does it well. The argument is narrower: it's about whether the default assumption of "put it in the cloud" is right for employee personal information, compensation data, performance evaluations, and disciplinary records. **Cloud makes sense when** you need real-time multi-user access across locations, complex integrations like payroll tax calculations, or when convenience clearly outweighs data sensitivity. **Local makes sense when** the data is highly sensitive, one or two people manage it, you don't need real-time collaboration, and you want to eliminate third-party risk entirely — especially if you're processing employee data through AI. For many small companies, HR is one or two people managing policies and employee records. They don't need the cloud. They need their data organized and accessible — on their own terms. ## The Control Question This comes down to control. Cloud means delegating control to a vendor — trusting them to secure your data, not misuse it, and give it back if you leave. Local means control stays with you, along with the security burden. You need to back up your machine, encrypt your drive, and practice good hygiene. But the blast radius of a local failure is limited to your own practices. A vendor breach can expose millions of records across thousands of companies. For the most sensitive category of business data — information about real people's lives, health, performance, and compensation — keeping that control isn't paranoia. It's prudence. ## What This Looks Like in Practice Going local-first doesn't mean going back to filing cabinets. The practical approach: 1. **Keep payroll in the cloud.** Payroll requires bank integrations and tax calculations that genuinely need cloud infrastructure. 2. **Move sensitive HR records local.** Employee files, performance docs, compensation data, and policies can all live on your machine. 3. **Use local AI for HR tasks.** Instead of pasting employee data into cloud AI, use local models that process everything on your hardware. 4. **Back up encrypted.** Local-first doesn't mean no backups. Use encrypted local or external drive backups. The goal isn't eliminating cloud. It's being intentional about which data goes where. --- [People Partner](https://peoplepartner.io) is a local-first HR knowledge platform that runs entirely on your Mac. Your employee data, policies, and documents stay on your machine — processed by on-device AI, never uploaded to the cloud. $99 one-time purchase. --- # Best HR Software with No Monthly Fees > Most HR tools charge per-seat monthly. Here's what exists if you want to pay once and own your software. *Canonical URL: https://peoplepartner.io/blog/hr-software-no-monthly-fees* *Published: 2026-02-24* *Keywords: hr software no subscription, hr software one time purchase, hr software no monthly fees* ## The Per-Seat Problem Most HR software charges you monthly. Per seat. Forever. At $8/employee/month, a 20-person company pays $1,920 per year. At $15/employee/month — common for mid-tier HRIS platforms — that's $3,600. Add another tool for performance management, another for surveys, another for document management, and you're easily north of $5,000 annually. For a funded startup burning through cash, that's a rounding error. For a bootstrapped company, a nonprofit, or a small business watching every dollar, it's a real expense — especially because it only goes up as you hire. The subscription model works great for software vendors. For buyers? It depends on what you're getting for the money. ## Why Most HR Software Is Subscription-Based It's worth understanding why the landscape looks this way before exploring alternatives. **SaaS is the default business model.** Since the mid-2010s, nearly all business software has moved to subscription pricing. It provides vendors with predictable recurring revenue, funds continuous development, and allows for cloud hosting costs. **Cloud infrastructure costs money.** If your data lives on someone else's servers, someone has to pay for those servers. Monthly fees cover hosting, security, backups, and uptime guarantees. **Continuous updates are expected.** When compliance rules change, subscribers expect the software to update. That ongoing development costs money, and subscriptions fund it. These are legitimate reasons. Subscription HR software isn't a scam — it's a business model that makes sense for certain products and certain customers. But it's not the only option. ## When Subscriptions Make Sense Let's be fair. Monthly HR software is the right choice when: - **You need payroll processing.** Payroll involves bank integrations, tax calculations that change quarterly, and regulatory compliance that varies by jurisdiction. This genuinely requires ongoing infrastructure and updates. Gusto, Rippling, and similar tools earn their subscription. - **You need benefits administration.** Managing health insurance, 401k, and other benefits involves real-time integrations with carriers and brokers. This is hard to do without cloud infrastructure. - **You need multi-user collaboration in real-time.** If 10 managers need to simultaneously access and update employee data from different locations, cloud-based tools handle this cleanly. - **You're scaling rapidly.** If you're hiring 5 people a month and need applicant tracking, onboarding workflows, and automated compliance at scale, the full HRIS suite pays for itself in time saved. ## When Subscriptions Don't Make Sense Monthly fees are harder to justify when: - **You're a small team (under 25).** The features you're paying for are designed for companies 10x your size. You're using 15% of the platform. - **Your needs are stable.** You're not changing your PTO policy every month. You don't need real-time dashboards updating by the minute. Your HR questions are important but not high-frequency. - **You value ownership.** With subscriptions, you're renting access. If you stop paying, you lose access to your own data (or at best, get a CSV export). Some organizations — especially those handling sensitive employee data — prefer to own their tools outright. - **Budget predictability matters.** Per-seat pricing means your costs grow with headcount. That's fine if revenue scales proportionally. It's painful if it doesn't. ## The Alternatives: HR Software Without Monthly Fees Here's what actually exists outside the subscription model. ### Open-Source HR Software **OrangeHRM Community Edition** The most established open-source HRIS. Covers employee records, leave management, time tracking, and basic recruitment. You self-host it, which means you need technical knowledge (or a developer) to set up and maintain it. - Cost: Free (self-hosted) - Pros: Full control, no per-seat fees, active community - Cons: Requires server setup, maintenance burden, UI is dated, limited support without paid plan - Best for: Companies with technical staff who can manage a self-hosted application **Odoo HR Module** Odoo is a full business suite with an HR module. The Community Edition is open-source. It handles recruitment, appraisals, time off, and employee records. - Cost: Free (Community) or one-time hosting setup cost - Pros: Integrated with other business tools (accounting, project management), customizable - Cons: Steep learning curve, the full suite can be overwhelming, Enterprise features require subscription - Best for: Companies already using Odoo for other business functions **IceHRM** A simpler open-source HRIS focused on small to mid-size companies. Employee management, leave tracking, time tracking, and basic reporting. - Cost: Free (self-hosted) - Pros: Simpler than OrangeHRM, cleaner interface, Docker deployment available - Cons: Smaller community, fewer integrations, limited documentation - Best for: Small companies wanting a lightweight self-hosted solution ### One-Time Purchase Software **People Partner** A local AI-powered HR knowledge platform for Mac. It consolidates your HR documents, policies, and employee data into a single searchable, AI-powered knowledge base. Runs entirely on your machine — no cloud, no servers, no ongoing costs. - Cost: $99 one-time - Pros: AI-powered search across all your HR documents, no subscription, data stays local, no technical setup - Cons: Mac-only, focused on knowledge management rather than full HRIS features (no payroll, no benefits admin) - Best for: Small teams and accidental HR managers who need fast answers from their existing HR documents ### Freemium Tiers (Free With Limits) Some subscription-based platforms offer genuinely useful free tiers: **Zoho People Free Plan** — Up to 5 employees. Basic employee database, leave tracking, and time tracking. **Bitrix24 Free Plan** — Up to 12 users. Includes HR tools alongside CRM and project management. **Freshteam Free Plan** — Limited to 50 employees for basic HR features. (Note: check current availability, as free tiers sometimes change.) The catch with freemium: features are limited, and you'll hit upgrade walls at exactly the moment you need more. They're designed to get you in the door. ## A Practical Decision Framework Choosing the right approach depends on three factors: ### 1. What Do You Actually Need? Be honest about this. List the HR functions you perform weekly. Not the features that sound nice — the ones you'd actually use. If your list is "answer policy questions, store employee docs, track PTO, and document performance conversations," you don't need a full HRIS. ### 2. What's Your Technical Comfort Level? Open-source solutions are powerful but require setup and maintenance. If you don't have someone comfortable with server administration, self-hosting adds more problems than it solves. One-time purchase desktop apps and freemium cloud tools require less technical overhead. ### 3. What's Your Real Budget? Calculate the 3-year total cost, not just the monthly price. - **Subscription at $10/seat/month, 15 employees:** $5,400 over 3 years - **Open-source self-hosted:** $0-500 (hosting costs) over 3 years - **One-time purchase:** $99 once The cheapest option isn't always the best, but understanding the real numbers helps you make an informed choice. ## The Honest Take There's no single answer here. If you need payroll and benefits administration, you're going to pay monthly — and that's fine. Those services genuinely cost money to operate. But if your core need is HR knowledge management — answering questions, storing policies, documenting decisions, and having a reliable reference — you have options that don't require an indefinite financial commitment. The subscription model has trained us to think renting software is the only way. It isn't. For many HR functions, especially at smaller companies, owning your tools makes more sense. --- [People Partner](https://peoplepartner.io) is a one-time purchase ($99) HR knowledge platform that runs locally on your Mac. No subscription, no cloud dependency — just your HR data, organized and searchable with AI. --- # HR Software for the Accidental HR Manager > You were hired to recruit. Or run ops. Or build the product. But now you're also the entire HR department. Here's what actually helps. *Canonical URL: https://peoplepartner.io/blog/accidental-hr-manager* *Published: 2026-02-21* *Keywords: accidental hr manager, hr tools for founders, hr software for small teams* ## You Didn't Apply for This Job One day you're the operations lead. Or the co-founder. Or the office manager. Then the company hits 8 employees, and suddenly you're fielding questions about PTO policies, performance reviews, and whether you need an employee handbook. Nobody hired you to do HR. There was no transition. No training. Someone just started CCing you on the sensitive stuff, and now here you are, Googling "how to write a PIP" at 10pm on a Tuesday. Welcome to accidental HR. It's one of the most common roles in small companies, and one of the least supported. ## The Trigger Moments Accidental HR doesn't happen gradually. It happens in moments. Specific, often uncomfortable moments. **The first performance problem.** Someone isn't working out. You've never fired anyone before. You don't know what documentation you need, what's legally required in your state, or how to have the conversation without exposing the company to risk. **The first policy question you can't answer.** "Do we have a bereavement leave policy?" You don't know. You're not even sure you need one yet. But someone's parent just died, and they need an answer today. **The first compliance scare.** You get a letter from your state labor board. Or someone mentions "I-9 audits" at a founder meetup and you realize you've never completed one correctly. The pit in your stomach is real. **The first sensitive disclosure.** An employee tells you they're pregnant. Or dealing with a disability. Or experiencing harassment from a coworker. Suddenly you need to know FMLA, ADA, and Title VII — and you need to know them now. These moments are when accidental HR managers realize: winging it isn't going to work anymore. ## What You Actually Need (And What You Don't) Here's what most HR software vendors would tell you: you need a full HRIS with applicant tracking, benefits administration, payroll integration, time tracking, performance management modules, and a learning management system. That's absurd for a 12-person company. Here's what you actually need: ### The Essentials - **A handbook.** Not a 60-page corporate document. A clear, concise set of policies that cover the basics: PTO, sick leave, anti-harassment, at-will employment, remote work expectations. Templates exist. Use one. - **Quick answers to HR questions.** When an employee asks about jury duty leave or you need to know the overtime rules in California, you need a reliable answer in minutes, not hours. - **A place for employee records.** Start dates, roles, compensation history, signed documents. Doesn't need to be fancy — but it needs to exist and be organized. - **Performance documentation.** Notes from 1:1s, feedback, any formal reviews. This is the stuff that protects you legally and helps you make fair decisions. ### The "Nice to Have Eventually" List - Formal onboarding workflows - Benefits administration - Applicant tracking - Automated compliance reminders - Org chart and reporting structure ### What You Can Skip Entirely (For Now) - Enterprise HRIS platforms (BambooHR, Workday, etc.) — overkill under 25 employees - Learning management systems — use Google Docs - Complex analytics dashboards — you don't have enough data for them to be meaningful - Any tool that requires a 3-month implementation ## The Real Problem: You Don't Know What You Don't Know The hardest part of accidental HR isn't the volume of work. It's the uncertainty. You're making decisions with real consequences — legal, financial, emotional — and you're not sure if you're doing it right. Every HR situation feels like it could be the one where you accidentally violate employment law. And the scary part? You're often right to worry. Employment law is complex, varies by state, and changes regularly. The difference between handling a termination correctly and incorrectly can be a lawsuit. This is why "just Google it" isn't a real strategy. Google gives you 10 different answers from 10 different states, half from 2019, and none specific to your situation. What accidental HR managers need is **contextual, reliable guidance** — answers that account for their specific policies, their state's requirements, and the actual situation they're dealing with. ## Building Your Minimum Viable HR Function If you're the accidental HR manager, here's a practical roadmap. It's not comprehensive. It's not perfect. But it'll keep you out of trouble and give your employees what they need. ### Month 1: Get the Basics Down - **Write (or adopt) a basic handbook.** Use a template from your state's labor department or a reputable HR site. Cover the must-haves: at-will employment, anti-discrimination, PTO, sick leave, and workplace conduct. - **Create an employee file system.** One folder per employee. Include offer letter, signed handbook acknowledgment, W-4, I-9, and any other signed documents. - **Learn your state's requirements.** Every state has different rules. Know the big ones: minimum wage, overtime, meal breaks, required postings, and mandatory policies. ### Month 2: Build Your Response Playbook - **Document your policies** so you can give consistent answers. If three people ask about remote work, they should all hear the same thing. - **Create templates** for common situations: verbal warning, written warning, PIP, termination checklist. - **Find a reliable resource** for HR questions. This could be an HR consultant on retainer, a platform with state-specific guidance, or an AI tool trained on employment law. ### Month 3: Start Thinking Ahead - **Set up a review process.** Even informal quarterly check-ins are better than nothing. Document them. - **Audit your compliance.** Are your I-9s complete? Do you have required workplace postings? Are you classifying employees vs. contractors correctly? - **Plan your handoff.** If the company is growing, you'll eventually need a real HR person. Start documenting everything so the transition is smooth. ## The Mindset Shift The biggest mistake accidental HR managers make is trying to build a Fortune 500 HR department at a 12-person company. You don't need that. You need just enough structure to: 1. Treat employees fairly and consistently 2. Stay compliant with employment law 3. Document decisions so you can defend them 4. Answer questions quickly and accurately That's it. Everything else is optimization. The second mistake is trying to keep everything in your head. You can't. Write things down. Document decisions. Save emails. Create a paper trail — not because you're paranoid, but because memory is unreliable and turnover is real. ## You Don't Have to Figure This Out Alone Being the accidental HR manager is isolating. You can't talk to employees about their own HR issues. You can't always talk to your co-founder about sensitive situations. And you definitely can't post the details on Reddit. Find a peer group, an HR advisor, or a tool that can serve as your sounding board. The worst thing you can do is make decisions in a vacuum. --- [People Partner](https://peoplepartner.io) was built for exactly this situation — the person who's doing HR but wasn't hired for HR. It's a local AI-powered knowledge base that gives you quick, reliable answers to HR questions based on your own policies and documents. Runs on your Mac, $99 one-time, no subscription. --- # How to Build an HR Knowledge Base from Scattered Tools > Your employee data lives in 5 different tools. Here's how to pull it all together into one place so you can actually make informed HR decisions. *Canonical URL: https://peoplepartner.io/blog/hr-knowledge-base-scattered-tools* *Published: 2026-02-18* *Keywords: hr knowledge base, organize employee data, hr data management* ## The Problem Nobody Talks About You have employee data. Plenty of it. What you don't have is employee data you can actually *use*. It's in Google Sheets. It's in your HRIS. It's in performance review exports. It's buried in a shared Google Drive folder someone created two years ago. It's in Slack threads you'd rather forget. It's in survey results sitting in a Typeform dashboard no one checks anymore. And every time someone asks a straightforward question — "What did we discuss in Sarah's last review?" or "When does our parental leave policy apply?" — you spend 15 minutes hunting across three different systems. This isn't a data problem. It's a retrieval problem. And it's costing you more than you realize. ## Why Scattered Data Leads to Bad HR Decisions When information is hard to find, people stop looking for it. That's human nature. But in HR, the consequences are real. A manager makes a termination decision without checking the employee's full performance history. Someone gives policy guidance based on a version of the handbook from 2023. A new hire's accommodation request gets lost between email and the ticketing system. None of these are malicious. They're the predictable result of a system where no one can find anything. **The hidden costs of scattered HR data:** - **Inconsistent answers.** Two employees ask the same policy question, get different responses because the person answering found different source documents. - **Slow response times.** Simple questions take 10-20 minutes instead of 30 seconds because you're searching across platforms. - **Compliance gaps.** Documentation exists but isn't connected to the people it applies to. - **Repeated work.** You write the same guidance email over and over because there's no central reference. ## What an HR Knowledge Base Actually Is An HR knowledge base isn't a fancy database. It's not an enterprise platform with dashboards and role-based permissions and a six-month implementation timeline. At its core, it's one place where you can find everything about your people and your policies. That's it. It should answer two types of questions: 1. **Policy questions:** "What's our PTO policy for part-time employees?" or "How do we handle requests for remote work?" 2. **People questions:** "What's the history of feedback for this person?" or "What training has this team completed?" If you can answer both types in under a minute, you have a functioning knowledge base. The format almost doesn't matter. ## How to Build One (Practically) You don't need to buy software first. You need to do an audit first. ### Step 1: Map Where Your Data Lives Spend 30 minutes listing every tool and location that holds employee-related information. Be thorough. Common places people forget: - Email threads with policy decisions - Slack channels where HR questions get answered - Personal notes from 1:1s - Onboarding checklists in project management tools - Compensation data in finance spreadsheets - Signed documents in DocuSign or HelloSign The goal is a complete inventory. You'll probably find 8-12 sources, even at a small company. ### Step 2: Categorize by Type Group your sources into categories: - **Policies and handbooks** — the rules - **Employee records** — the people data - **Performance history** — reviews, feedback, goals - **Compliance docs** — signed agreements, certifications, I-9s - **Institutional knowledge** — past decisions, precedents, context This tells you what you're working with and what gaps exist. ### Step 3: Choose a Consolidation Strategy You have three realistic options: **Option A: The "Good Enough" Approach** Create a well-organized folder structure (Google Drive, Notion, whatever your team uses). One folder per category. Naming conventions enforced. A master index document that tells people where to find things. Cost: Free. Time: A weekend. Limitation: Still requires manual searching. **Option B: The Wiki Approach** Use a tool like Notion, Confluence, or Slite to build a searchable wiki. Import your policies, create templates for employee records, link related documents. Cost: $8-15/user/month. Time: 1-2 weeks. Limitation: Requires ongoing maintenance, and search is only as good as your tagging. **Option C: The AI-Powered Approach** Feed your documents into a system that can understand them, connect them, and answer questions about them. This is where things get interesting — you're not just organizing files, you're creating a layer that can reason across your entire HR knowledge. Cost: Varies. Time: Hours to set up. Limitation: Quality depends on the tool. ### Step 4: Establish Maintenance Habits A knowledge base that isn't maintained becomes a graveyard. Set two habits: 1. **Update on change.** When a policy changes, when someone's role changes, when a review is completed — update the knowledge base immediately. Not later. Now. 2. **Quarterly audit.** Once a quarter, scan for outdated information. It takes 30 minutes and prevents the slow decay that makes knowledge bases useless. ## The Real Unlock: Being Able to Ask Questions The biggest shift isn't having organized files. It's being able to ask a question in plain language and get a reliable answer. "What did we decide about remote work for the engineering team?" Instead of searching four tools, you get an answer that synthesizes the policy, the team-specific exception, and the relevant Slack conversation where leadership made the call. That's the difference between a filing cabinet and a knowledge base. One stores things. The other understands them. This matters most when you're making decisions under pressure — handling a sensitive employee situation, responding to a compliance question, or advising a manager who needs guidance now, not after you've spent an hour digging through archives. ## Start Small, but Start You don't need to solve this in a week. Pick your most painful data silo — the one that causes the most daily friction — and consolidate that first. For most teams, it's either policy documents or performance history. Get one category right. Then expand. The goal isn't perfection. It's having one place you trust when someone asks you a question. Everything else builds from there. --- If you want to skip the manual consolidation and go straight to an AI-powered HR knowledge base, [People Partner](https://peoplepartner.io) pulls your HR documents, policies, and employee data into a single local app that runs on your Mac. One-time purchase, no cloud dependency, and your data stays on your machine.