OptimisedI once inherited a team where the check-in system sent a Friday afternoon email that read, in full, "Please rate your week from 1 to 5." Nobody replied to it for eleven months, and the manager who set it up truly believed she was running continuous performance management. That experience taught me the rule I now apply everywhere I work: automate the plumbing, never the conversation.
If you want the short version, here it is. You can automate the scheduling, the reminders, the agenda skeleton, the record keeping and the follow up on actions. You cannot automate the judgement, the questions that matter, or the moment where a person tells you something is wrong. Get that split right and automation makes check-ins better, not colder. Get it wrong and your team learns to ignore the robot within a month, exactly as mine did.
This guide is written for the UK, so the law, the tools and the prices are British, with a couple of American statistics flagged as such where they're the only credible numbers. I'll show you what to automate, what to leave alone, which platforms I'd pay for, and how to stay on the right side of Acas and the Information Commissioner's Office.
Why Check-Ins Go Robotic In The First Place
Most of the mess comes from a category error. A check-in is a conversation between two people. A workflow is a sequence of tasks. When you buy performance management software, the vendor sells you a workflow, and the temptation is to let the workflow become the conversation. That's how you end up with managers filling in a form the night before, reading it aloud, and calling it a one-to-one.
The second cause is frequency without purpose. The research is fairly clear that regular conversations beat an annual appraisal. The CIPD's own senior adviser on organisational behaviour, Jonny Gifford, has said the evidence supports performance management being done more continually, as ongoing conversations rather than a once or twice a year event. But "more often" is not a design. If a fortnightly check-in has no agenda beyond "how's it going", people stop preparing and within a quarter the meeting is a calendar block both sides quietly hope gets cancelled.
The third cause is that automation is easy to buy and hard to shape. Every UK platform I looked at will happily fire off reminders, chase completions and generate reports. None of them will tell you what a good question sounds like. That part is still on you.
What The Numbers Actually Say About Frequency
Let me give you the honest evidence picture, because a lot of what gets quoted in this space is marketing dressed up as research.
The most useful UK source is the CIPD's evidence review on performance feedback, written by Ioana Cioca and Jonny Gifford in 2022. Its headline finding is not that more feedback is always better. It's that feedback is generally a positive method for improving performance but can be detrimental if it's not done in the right way. On the specific question of promptness, the authors are careful: they say workplace research on whether feedback benefits from being prompt is limited, though it gives some support. That's a far more measured claim than "weekly check-ins boost engagement by 40 percent", which is the kind of line you'll see on vendor landing pages.
On the employer side, Acas commissioned NatCen Social Research to look at how UK organisations actually run performance management. As Menzies Law summarised when the Acas guidance was published, one in ten employees felt their employer's system was a source of demotivation, and only one in four organisations adapted their processes for staff with disabilities or conditions such as dyslexia and autism. That second figure matters more than it looks, and I'll come back to it under the law section.
The American figures are starker, and I'll flag them as American. PerformYard's 2025 State of Performance Management report found that 95 percent of managers were dissatisfied with their current performance management system, according to the summary published by Speakwise. Gallup's research, as reported by EvalFlow, found only 47 percent of employees strongly agreed they knew what was expected of them at work, and that employees who reviewed progress with their manager quarterly were 90 percent more likely to be engaged. These are US samples, so I wouldn't quote the exact percentages to a UK board as if they were ours, though the direction of travel matches what I've seen here. Avado, a CIPD accredited training provider, does cite CIPD research suggesting over 40 percent of UK employees report not having clear performance expectations, which lines up closely enough with the Gallup figure to suggest the problem isn't confined to one side of the Atlantic.
The takeaway is simple. Frequency helps, clarity helps more, and a badly run frequent check-in does active harm. Point automation at clarity and consistency, not at cadence for its own sake.
The Split That Keeps Check-Ins Human
Here is the framework I use, and it's the one thing I'd ask you to take away even if you skip the rest of the article. Divide every part of a check-in into one of three buckets.
Fully automate: scheduling and rescheduling, calendar invites, reminders to both parties, pulling in goal progress and any peer feedback since the last meeting, storing the notes, and nudging on unfinished actions. This is admin. No human should be spending Monday morning on it.
Automate the scaffold, not the content: the agenda. A good tool pre populates a shared document with a light structure, but the manager and the employee write into it. The structure I favour has four lines. What went well since last time. What got in the way. What I need from you. One thing for next time. That's it. Four prompts, no ratings, no sliders.
Never automate: the actual questions a manager asks in the room, the decision about whether something is a development issue or a capability issue, anything that feeds into pay or promotion, and any conversation about wellbeing. If your software offers an AI summary of "employee sentiment" from check-in notes, I'd switch it off for now, and the law section explains why.
This split removes the friction that causes managers to skip check-ins without removing the parts that make them worth having. In my experience the biggest predictor of whether one-to-ones survive a busy quarter is whether the manager has to do anything to make the meeting happen. If the invite, the reminder and the notes doc all appear on their own, the meeting happens. If any one of them needs a manual step, it slides.
Setting The Cadence
I'd default to a 30 minute one-to-one every two weeks for most knowledge work roles, a monthly check-in for experienced people in stable roles, and weekly for anyone in their first 90 days. The CIPD community discussion on appraisal good practice includes a practitioner describing how their organisation set a minimum of one check-in every four weeks after moving off annual appraisals, and deliberately didn't police it beyond that, letting people find what worked. I think that's the right instinct. Set a floor, not a ceiling, and let teams go above it.
Automation earns its keep by enforcing the floor without nagging. A good setup sends one reminder to the manager two working days before, one to the employee the day before with a link to the shared agenda, and an escalation to HR only if a check-in has been skipped twice in a row. Three touches, not thirty.
New starters deserve a separate track. Factorial's comparison of UK performance management software notes that BambooHR lets you schedule and automate reminders for 30, 60 and 90 day check-ins with new hires. That pattern is worth copying regardless of platform, because the first three months are exactly when a missed conversation does the most damage.
The Tools I'd Actually Use In The UK
I'm going to be opinionated here, with real prices. Every figure below is the published price at the time of writing, exclusive of VAT unless stated, and I'd expect them to drift upward over the next year.
Breathe. Built in the UK, owned by ELMO Software, and priced per business rather than per head, which is unusual and useful for small teams. Breathe's own pricing page says plans start at £24 per business per month, and the independent Breathe HR review by ExpertSure lays out the full ladder: Micro at £24 a month for 1 to 10 employees, Starter at £44 for 11 to 20, Regular at £99 for 21 to 50, Pro at £175 for 51 to 100, Premium at £419 for 101 to 150 and Plus at £579 for 151 to 200, with around 15 percent off for annual billing on the smaller plans. Performance management is included in the core plan, covering review cycles, goals and one-to-one agendas. It doesn't do 360 feedback and its analytics are thin, so if you're 150 people and want calibration tools, look elsewhere. Best for: UK businesses under 50 people who want check-in reminders and a notes record without paying per seat.
CharlieHR. Another UK native, aimed squarely at startups and sub 50 employee firms. According to the CharlieHR review published by ExpertSure, the Core plan starts at £20 per month for teams of 1 to 4 and scales to £735 a month at 250 plus headcount, with a 75 percent discount for the first six months on the smallest tier, which brings it to £5. CharlieHR's own pricing page confirms all prices exclude UK VAT at 20 percent and that performance reviews are part of the core feature set. It's clean and quick to set up. It's also limited past 50 people, with basic reporting and no payroll. Best for: a founder who wants one-to-ones happening reliably across a team of 10 to 30 with almost no admin.
PeopleHR and Moorepay. Both are UK payroll and HR suites with performance modules bolted on. PeopleHR advertises flexible check-in cadences with automated reminders and criteria tailored by role. Moorepay's pitch is explicitly light touch: automated pings to managers, nudges on submission, and reporting on who has and hasn't completed a review. Neither publishes a list price, so you'll be talking to sales. Best for: organisations already on one of these for payroll who want check-ins in the same system rather than a separate login.
Actus. A UK performance management specialist rather than a general HR suite. It does automated calendar invites for reviews, scheduling and documenting of one-to-ones, and real time dashboards showing whether performance conversations are actually happening. Again, pricing is by quote. Best for: HR teams of 100 to 1,000 who want performance as the primary product rather than an add-on.
15Five. This is the American platform with the most transparent pricing, and it bills in US dollars, so every pound figure I give for it is approximate. According to Pulsewise's comparison of engagement software pricing, the published plans updated in February 2026 are Engage at $4 per user per month, Perform at $11, and Total Platform at $16, on annual billing. That's roughly £3, £8.50 and £12.50 per person per month at the time of writing. G2 reviewers, as summarised on the platform's comparison page, single out 15Five's weekly check-ins as a strength, describing them as keeping teams aligned without feeling micromanaged. I'd agree with that. Best for: a UK team that wants the lightest weekly check-in ritual and doesn't mind a dollar invoice.
Lattice. The other big American name. Engagedly's comparison, drawing on Lattice's official pricing page, lists Performance at $10 per seat per month and the Foundations bundle at $13, with a $4,000 minimum annual agreement. That minimum is roughly £3,100 and it makes Lattice a poor fit for anything under about 30 people. Lattice is excellent at OKRs, which stands for Objectives and Key Results, a goal setting method where a broad objective is paired with two or three measurable results. If you run OKRs it's hard to beat. If you don't, you're paying for a lot you won't use. Best for: 50 plus person companies with a mature goal setting culture and a US dollar card.
Leapsome. I'd be cautious. The modules are quote gated, starting prices vary wildly across review sites, and Pulsewise reports a $6,000 per year minimum on its customer success tier. If you can't get a straight price before the demo, expect a hard renewal conversation in year two.
A Workflow You Can Set Up This Week
This is what I'd build on Breathe or CharlieHR for a team of 20, and it takes about two hours.
Step one: create a recurring fortnightly 30 minute one-to-one for each manager and report, owned by the report. The report owns it because the meeting is for them. When the manager owns it, it gets cancelled for the manager's convenience.
Step two: set up a shared note template with the four prompts from earlier. What went well, what got in the way, what I need from you, one thing for next time. Both people fill in their side before the meeting. Neither fills in more than five lines per prompt. The point is to walk in with the same picture, not to write an essay.
Step three: configure two reminders. The report gets one 24 hours before with a link to the template. The manager gets one 48 hours before that includes any goal progress and any peer feedback logged since the last meeting. That's the manager's only prep.
Step four: turn on action tracking. Every "one thing for next time" becomes a task with a due date. The tool nudges once at the due date and once a week later. After that it stops, and the item comes up in the next check-in instead. Automated nagging past two touches is how people learn to ignore the system.
Step five: set an HR alert for two consecutive skipped check-ins. Not one. Everyone misses one. Two in a row is a pattern, and the right response is a human conversation with the manager, not another automated email.
Step six: review the setup after 90 days with three questions. Are check-ins happening at the floor cadence. Are the notes written by both people or just one. Has anything in the notes surprised you. If the answer to the third is no, your template is too safe.
The Questions That Do The Real Work
I want to spend a moment on this because it's where automation stops and craft begins. Acas's guidance on reviews and appraisals is explicit that alongside formal reviews, managers should still talk to employees informally and regularly through feedback, coaching and check-ins, and that it's important to keep a written record that's shared with the employee afterwards. The tool handles the record. You handle the talk.
The questions I've found most useful in a fortnightly one-to-one are short and slightly uncomfortable. What's the thing you've been putting off. What did I do in the last two weeks that made your job harder. If you could change one decision I made, which one. What would you do with a free afternoon this week. None of these fit on a slider, and none of them should.
Avoid anything that sounds like a survey. "On a scale of one to ten, how supported do you feel" is a pulse survey question, and pulse surveys belong in a separate anonymous channel, not in a conversation where the person is looking their manager in the eye. Mixing the two makes check-ins feel like a compliance exercise.
Where Automation Meets The Law
This is the section most articles skip, and it's the one that can land you in front of an employment tribunal. The main bodies you need to think about are Acas, the Information Commissioner's Office, and the Equality Act 2010.
Start with data protection. The ICO published its guidance on employment practices and monitoring workers on 3 October 2023 to help employers comply with UK GDPR and the Data Protection Act 2018. UK GDPR is the British version of the European data protection regulation, retained after Brexit and enforced by the ICO. The guidance explicitly notes that employers are increasingly using data analytics to infer worker performance and wellbeing, and that any such monitoring must be lawful and fair. The ICO also gives a pointed example: an employer that assesses performance by monitoring how long workers spend in a case management system, without accounting for work done outside it, is carrying out monitoring that's unfair and inadequate. Translate that into check-in automation and it means any dashboard that scores people on completion rates or logged activity needs a human reading it with context.
The bigger risk is Article 22 of UK GDPR, which gives people the right not to be subject to a decision based solely on automated processing where that decision has legal or similarly significant effects. Farrer and Co's summary of what the ICO guidance means for employers makes the point that this covers not just dismissal but other significant workplace decisions, such as increasing or decreasing pay based on performance, and that employers must not disadvantage workers who ask for human intervention. So if your platform's AI produces a performance score, and that score feeds a pay decision without a manager actively reviewing and being able to override it, you have a problem. This is why I switch off AI sentiment summaries by default. It's not that they're useless. It's that they're a lawsuit waiting for a lazy manager.
One caveat on timing. The ICO's own monitoring page now carries a note that the guidance is under review because of changes made by the Data (Use and Access) Act, so check for updates before you build anything that leans heavily on automated processing. The principles haven't changed, but the detail might.
Then there's Acas. The Acas Code of Practice on Disciplinary and Grievance Procedures governs what happens when a check-in surfaces a capability problem and you move to a formal process. Your automated notes become evidence at that point, which is a good reason to make sure they're written by both parties and shared, exactly as Acas recommends. Acas's tips on getting performance management right also stress avoiding surprises by raising problems as they occur rather than saving them for the end of year meeting, and avoiding discrimination by ensuring arrangements are fair to everyone in line with the Equality Act 2010.
That last point connects to the one in four figure from earlier. If your check-in template is a dense written form and one of your team has dyslexia, an automated fortnightly demand to fill it in is not neutral. Reasonable adjustments apply to performance processes just as they apply to the job itself. A voice note option, a shorter template, or simply agreeing that the manager types while the employee talks are all cheap fixes, and they're exactly the sort of thing a rigid workflow can quietly prevent.
An Honest Reality Check
Automation will not fix a manager who doesn't want to have the conversation. I've watched beautifully configured systems produce two years of empty notes because the manager treated the meeting as a status update and the reports learned to give one. The Betterworks research cited by EvalFlow, again a US sample, found that two in three managers said they needed more support with performance management. In my experience the British figure isn't any better. If you're going to spend money, spend the first chunk on a half day of manager training on how to ask a question and then shut up, and the second chunk on software.
The second uncomfortable truth is that check-in data is seductive and mostly meaningless in aggregate. A completion rate of 94 percent tells you that 94 percent of meetings were logged, not whether anyone said anything true in them. The only aggregate I trust is the one I described in step six: did anything in the notes surprise you. If your HR dashboard can't answer that, don't let it drive decisions.
The third is that continuous performance management is a culture, and cultures don't ship with software. In the organisations that do this well, the most senior person in the building visibly keeps their own one-to-ones. When the chief executive cancels hers, everyone else's slide within a fortnight.
Deciding What To Do Next
Ask yourself four questions. Are your one-to-ones currently happening at all, or only when something goes wrong. If the honest answer is the second, start with scheduling automation and nothing else, and give it a quarter before adding templates. Is anyone on your team under 90 days in. If so, put them on a weekly track this week, regardless of what the rest of the team does. Does any decision about pay or promotion currently rely on a score the software generates. If yes, insert a human review step before next month's cycle, and document that the manager can override. And finally, has any employee asked for an adjustment to how their check-ins run. If nobody has, that's not evidence you don't need to offer one.
Then pick the tool that matches your size. Under 50 people, I'd go Breathe or CharlieHR and spend nothing else. Between 50 and 150 with an existing PeopleHR or Moorepay contract, add their performance module and negotiate hard. Above that, or if you run OKRs seriously, look at Actus or Lattice. Steer clear of anything that won't give you a price before a demo.
The thing to do this week is smaller than any of that. Open your own calendar, find the next one-to-one you've got with each person who reports to you, and write two lines into a shared note before it: what went well, and what got in the way. Send them the link and ask them to do the same. That's the whole system in miniature, and if it works by hand for a fortnight, you'll know exactly what you want the software to do when you automate performance check-ins properly.