The Problem
Enterprise goal-tracking lives inside systems like SuccessFactors. They are clunky and clumsy by design: built for HR compliance, not for the person whose goals they're supposed to hold. Doing anything takes ten clicks. It feels like work, so people avoid it, so it stays empty.
Then the worst part: the end-of-year scramble. Review season arrives and you're digging through a year of email, Slack, calendars, and memory trying to reconstruct what you actually accomplished. The thing that was supposed to track your year did nothing for eleven months, and now you're doing archaeology on your own work under deadline. It's a painpoint every single year, for everyone.
The goals aren't the problem. The container is. A goal you set in January is invisible by March because nothing ever brings it back up, and the record you need in December was never being kept.
The Idea
A bot that keeps track, understands context, and actively supports growth. The opposite of a form.
The name is the metaphor: a cairn is a stack of trail-marker stones. Each says you're on the path, here's how far you've come. Cairn stacks your wins the same way, in real time, so the trail behind you is always visible and the record is always current.
Log It The Second It Happens
The core move is frictionless capture. Something good happens, you log it on the fly: a sentence to the bot, a voice note, a quick tap. Not ten clicks through nested menus. One move, in the moment, while it's fresh.
And capture doesn't feel like data entry, because it feels like telling someone. Logging a win to Cairn is like sharing a success with a mentor or a colleague who's actually glad to hear it. It congratulates you, it cheers you on, and it adds something back: context, a connection to a goal, a "that's the third time this quarter you've unblocked the team, that's a pattern worth naming in your review." The encouragement is the point. People will share a win with someone who's rooting for them. Nobody fills out a form for fun.
Goals That Live Across The Year
Goals aren't set once and frozen. Through the year they get added as new priorities appear, amended as scope shifts, and archived when they're done or no longer relevant. The system holds the whole lifecycle, so the goal set always reflects reality instead of what someone typed in January.
Assessment As Navigation, Not Just Verdict
Ongoing assessment isn't only for review time. It keeps you on track. By watching the shape of your progress, Cairn can show where you're slipping before it becomes a problem and help you course-correct while there's still time. A goal going quiet, a streak breaking, momentum stalling on something that matters: the bot surfaces it, asks about it, and helps you adjust. Assessment becomes a steering input during the year, not a grade handed down at the end of it.
The End-Of-Year Payoff
Because everything has been compiled in real time, the year-end assessment is already done. No recon. No digging. No scramble. The bulletproof record exists at all times, ready for review on any given day, not reconstructed in a panic every December.
When review season comes, you open a document that's been writing itself all year: quick references to every logged win, charted achievements, quantifiable growth laid out as lines that climbed and bars that filled. The single most-hated week of the corporate year stops being a week of work and becomes a five-minute read.
The Reviewer's Side
The scramble isn't only the employee's problem. The manager doing the reviewing carries it too: a stack of direct reports, each a year of detail to wade through, all due at once. They're doing the same archaeology, multiplied by their team size.
Agent-to-agent assessment takes that strain off. The employee's Cairn agent has the compiled, evidence-backed record; the reviewer's agent can query it directly. Instead of reading eleven months of raw logs for every person, the reviewer asks their agent the questions that matter ("where did this person grow, where did they stall, what's the evidence") and gets a synthesized answer with the receipts attached, ready to verify rather than reconstruct.
Both sides get the same thing: the detail is there when you want to drill in, but nobody has to hold all of it in their head to do the review. The employee's agent makes the case; the reviewer's agent checks it. The human conversation that follows is about the work, not about who can remember what happened in Q1.
Fairer By Construction
Every stakeholder can dig into the same vast record for what matters to their point of view. The employee builds their case. The manager checks growth and gaps. A skip-level or HR can audit for consistency across a team. Peers can corroborate. Same evidence base, different questions, all grounded in what was actually logged when it actually happened.
That changes the failure modes of a review. Traditional performance reviews run on memory, and memory is where bias lives: recency bias (the last month outweighs the first eleven), halo effect (one good impression colors everything), similarity bias (people rate people like themselves higher), and the plain human error of forgetting what someone did in March. When the record is continuous and evidence-backed, the evaluation has to reckon with the whole year, not the most recent or most memorable slice. Decisions trace to logged events anyone can inspect. More transparency, more fairness, less room for the review to drift on who-remembers-what or who-the-manager-happens-to-like.
The honest caveat: grounding a review in evidence reduces human memory bias, but it doesn't make the system neutral by magic. The model doing the synthesis carries its own biases, and what gets logged is itself a choice. The win is auditability, not objectivity. Because every claim points to a verifiable event, a biased reading can be challenged with the record instead of argued from feelings. That's the real improvement: not a perfect judge, but a transparent one whose work can be checked.
Charting Growth
Progress rendered to satisfy. Humans like a line that climbs and a bar that fills all the way in, and Cairn leans into that completion drive on purpose: every goal has a bar, every bar wants to be full, watching it fill is its own reward. Streaks, milestones, a portfolio view of everything climbing at once. The charts double as the proof: quantifiable growth you can point at, not adjectives you have to argue for.
A Game You're Actually Playing
Somewhat gamified, on purpose, and somewhat is the operative word. Not badges and confetti and a cartoon mascot. The game mechanics that actually fit professional growth: a skill tree that fills in as you develop, levels that correspond to real capability, a sense that you're progressing through something rather than logging into something.
Growth reframed as a learning experience you're playing through. The course-correction nudges become the game telling you you've wandered off the path. The charts become your stats screen. The end-of-year record becomes your save file, the proof of the run. It stays professional, but it borrows the one thing games are better at than any productivity tool: making you want to come back tomorrow and move the number.
The "somewhat" is the discipline. Too much and it's insulting in a workplace. Just enough and the most tedious part of a job (documenting that you did it) becomes the part with a little pull to it.
How It Works
- Frictionless capture. Text, voice, or one-tap logging from wherever you are. The bot does the structuring after the fact, so capture stays effortless.
- Goal model. Each goal is structured: objective, milestones, next action, status, deadline, dependencies, a running log of wins and check-ins, and a lifecycle state (active / amended / archived). A thing the system can reason about and chart, not a free-text box.
- The mentor layer. An LLM with a coaching persona and the user's full history in context, trained in actual motivational technique: motivational interviewing, goal-setting theory, implementation intentions, positive reinforcement timing. Not generic cheerleading. It runs the follow-ups, asks the next-step questions, reframes stuck goals, and delivers encouragement that's grounded in what actually moves behavior. Tone is supportive colleague, not manager.
- The trigger engine. A scheduler watching for deadline proximity, staleness, milestone completion, slipping momentum, and period boundaries. Fires the proactive outreach and the course-correction nudges.
- The charts + the living assessment. Progress bars, achievement timelines, streak counters, and a continuously-compiled review document that's always current.
- Integration path. Sit alongside an existing system (read/write SuccessFactors, add the engagement layer on top) or replace the goal-tracking function outright for a team free to choose.
Why It's Different
The incumbent is built for the organization. Cairn is built for the person. The org wants a record at review time; the person wants something that helps them get the thing done and remembers it for them so the review writes itself. Those aren't the same product, and pretending they are is why everyone hates the form.
The wedge is emotional, not feature-based. People don't engage with a tracker. They engage with something that's glad when they win, keeps them honest when they slip, and quietly builds the case for them while they work.
Win, win, win. The employee gets a coach that makes the worst week of the year disappear. The reviewer gets a fair, auditable record instead of a memory test. The organization gets honest, evidence-backed development data and people who actually grow. Three parties, one system, and for once none of them is paying for the others' convenience.
Status
Concept, work-context. The pieces are standard (frictionless capture, structured goal model with lifecycle, an LLM coaching layer with history in context, a trigger scheduler, a charting front-end, a living-document compiler). The hard and interesting parts are the mentor layer's tone (encouraging without being saccharine) and the trigger engine's judgment about when to reach out, because a bot that pings too often becomes the next thing you mute. Pairs with Hive as the other work-context closed-loop proposal: Hive routes the team's tasks, Cairn climbs the individual's goals.