A team leader and a smiling agent with her headset around her neck share a short one-to-one at a desk, with a small timer beside them and a wall planner behind showing one coloured dot for each of four weeks.

Ask an operations director how often their agents are coached and the honest answer is often "not as often as we'd like". Most would add that it isn't for want of trying. A contact centre team leader runs the queue, covers absence, files the reports and is then expected to coach as well, and coaching is the one task on that list that can always move to next week.

So how often should it happen? We went looking for an official answer, in the industry standards and in the associations that represent BPOs and contact centres. What we found was a monthly floor, a surprising amount of silence, and a body of evidence that points somewhere quite different.

How often are agents coached today?

One of the clearest pictures comes from SQM Group, which benchmarks contact centre performance. Its research finds that in most cases four to five calls per agent are evaluated each month, and agents are coached once a month for less than an hour. Across a year, that is about a day and a half of coaching per agent (SQM Group, 2020).

The same research found a gap between what centres report and what happens. Many say their supervisors spend half their time or more coaching. SQM's measurements show it is much less (SQM Group, 2020).

McKinsey describes the same pattern. Coaching slots are often built into agents' schedules, but supervisors struggle to keep them as they juggle meetings, administration and special projects. And the coaching that does happen is often based on four to six calls out of the five or six hundred an agent handles each month (McKinsey, 2018).

None of this says team leaders don't care. It says the hours run out before the coaching does.

What do the standards and industry bodies ask for?

How often should contact centre agents be coached?

The formal standards set a monthly floor. ISO 18295 asks for each agent's contacts to be assessed monthly, COPC builds a monthly one-to-one around each agent's scorecard, and SQM Group recommends coaching every agent at least once a month for about an hour. Gallup's research points further, to a meaningful conversation of 15 to 30 minutes every week. The typical agent is coached once a month, for under an hour, using four or five calls.

What do ISO 18295 and COPC ask for on coaching?

ISO 18295-1, the international standard for customer contact centres, asks for an unbiased monthly sample of each agent's contacts to be assessed, a review of each agent's competencies at least once a year and continuous development. COPC makes the agent scorecard the main document in a monthly one-to-one, with more frequent check-ins for agents still learning. Neither sets a frequency for coaching conversations themselves.

ISO 18295 applies to in-house and outsourced centres alike (ISO, 2017). It has a South African root. In 2008 South Africa became the first country with national standards for running contact centres, SANS 990, and those became the foundation for ISO 18295, with BPESA supporting their development (CallForce, 2026).

COPC's standard is widely used by BPOs and the brands that hire them. Its guidance puts the agent scorecard at the centre of every monthly one-to-one between supervisor and agent, and recommends more frequent check-ins for agents who are learning the role or working towards a target (COPC, 2024).

SQM Group's view is that agents must be coached at least once every month, in a session of about an hour (SQM Group, 2022).

Does BPESA recommend a coaching frequency?

No. BPESA, South Africa's industry body for global business services, focuses on investment, jobs and skills. Its 2016 Key Indicator Report found 86.8 per cent of centres used side-by-side coaching between supervisor and agent to assess the quality of agents' communication, but it did not record how often centres coach. CXAP in the Philippines and the CCMA in the UK also publish no coaching frequency.

BPESA's 2018 report has no coaching data at all (BPESA, 2018; BPESA, 2016). CXAP, formerly the Contact Center Association of the Philippines, runs leadership forums, certification and capability-building programmes (CXAP). The CCMA accredits UK centres against its Contact Centre Standards Framework, which covers leadership, colleague experience and operational excellence (CCMA, 2026). None of them sets a frequency.

Put together, the standards ask centres to assess every month. None of them asks for a regular coaching conversation. A centre can meet all of them in full while its agents rarely sit down with their team leader to talk about their work.

What does the evidence say about frequency?

The strongest evidence on frequency comes from Gallup's long-running research on managers and engagement.

Gallup's conclusion is that the most important habit for a manager is a meaningful conversation with each person every week. These conversations don't need to be long. Between 15 and 30 minutes is enough, but only if it happens frequently, and short regular conversations have a greater impact than longer occasional ones. Managers who don't talk to people weekly need longer conversations to catch up (Gallup, 2023).

4x How much more likely employees who get weekly, meaningful feedback are to be engaged than those who don't Gallup, 2026
21% Employees who strongly agree they had meaningful feedback from their manager in the past week Gallup, 2026
70% The share of the variance in team engagement that Gallup attributes to the manager Gallup, 2026

The link between weekly feedback and engagement holds regardless of team size (Gallup, 2026). And managers often believe they are doing better than their people feel. Only 20 per cent of employees say they receive feedback weekly, while about half of managers say they give it (Gallup, 2024).

Gallup is specific about what makes a conversation meaningful. In order of importance, the topics are recognition for recent work, collaboration and relationships, current goals and priorities, and the person's strengths. The one topic employees see as less meaningful is a conversation about their weaknesses (Gallup, 2023). Only 16 per cent of employees say their last conversation with their manager was extremely meaningful (Gallup, 2026).

Gallup's research is worth reading with its source in mind. Gallup is a consultancy that also sells manager development, and its findings show strong associations rather than proof of cause. They do come from very large samples and have been consistent for many years, which is why they are widely used. For more on Gallup's latest data, see the global engagement crisis and engagement, satisfaction and happiness.

Where is the gap between the standards and the evidence?

Set the three pictures beside each other.

  • What the standards require: a monthly assessment of each agent's calls, with a yearly competency review.
  • What the evidence supports: a weekly conversation of 15 to 30 minutes about recognition, goals, colleagues and strengths.
  • What typically happens: a monthly session of under an hour, built from four or five calls, held when the queue allows.

The rhythm Gallup describes is about four times what the standards require. And the two measure different things. The standards ask whether calls were scored. Gallup asks whether a person had a conversation that mattered to them.

The gap matters most at the start. COPC finds that contact centres typically see around 70 per cent or more of their first-year attrition within the first ninety days (COPC, 2024), and SQM Group reports the same pattern, with the first ninety days being when most agents quit (SQM Group, 2023). That is the window in which a new starter is finding out whether anyone here understands how they work. On a monthly rhythm, a new agent has two or three coaching conversations before most of that decision is made. We look at this in more detail in why new call centre agents leave in the first 90 days.

Is more coaching always better?

It would be easy to read all this and conclude that the answer is simply more coaching. The evidence is more careful than that.

Does more coaching always help?

Not on its own. Gartner's research with 7,309 employees found no significant relationship between the time a manager spends developing people and how well they perform. Managers who gave constant feedback across every skill lowered performance by up to 8 per cent, while managers who tailored their coaching to the person raised it by up to 26 per cent. Frequency helps when each conversation is meaningful and suited to the person.

Gartner calls the first group "always-on" managers and the second "connector" managers (AHRI, 2018, reporting Gartner). Employees who report to effective manager coaches are 40 per cent more engaged and 20 per cent more likely to stay (Gartner).

This sits comfortably with Gallup. Gallup's finding is about meaningful conversations, and Gartner's is about volume. Read together, they describe a regular rhythm of conversations that land, which is different from both the monthly session and a constant stream of feedback.

A landmark review of feedback research makes the same point from another angle. Kluger and DeNisi found that feedback made performance worse in over 38 per cent of cases, and that it tended to fail when it moved the person's attention from the task to how they were being judged (Kluger and DeNisi, 1996). We explore this in the last mile of coaching and the feedback sandwich, reviewed.

In practice, that suggests a predictable rhythm, one focus for each conversation, recognition first, and one agreed action that is followed up next time.

Why does coaching slip to monthly?

If team leaders know coaching matters, why does it settle at monthly? Most of the reasons are about time and visibility.

Preparation takes longer than the conversation

Before a team leader can coach, they have to find a call worth discussing, listen to it, score it against the campaign standard and plan what to say. Then they hold the conversation and write it up.

As an example, say it takes 15 minutes to find and listen to a call, 10 to score it, 20 for the conversation and 10 to write it up. That is 55 minutes per agent. For a team of 15 agents coached weekly, it comes to almost 14 hours a week, more than a third of a working week before the queue, absence and reports. Monthly is what the hours allow.

Other work competes for the same hours

Coaching slots are often in the schedule, but supervisors struggle to keep them as meetings, administration and special projects compete for the same time (McKinsey, 2018). Unless coaching is planned into the workforce schedule as protected time, more urgent work tends to take its place.

Coaching rests on a small sample of calls

When coaching draws on four to six calls out of several hundred, the agent can fairly wonder whether the call in front of them is typical of their work.

Few centres can see who has been coached

Most centres can report QA scores for every agent. Far fewer can say who had a coaching conversation this week. Without that view, a monthly rhythm looks normal, and the gap Gallup describes between what managers believe and what employees experience stays out of sight.

The same feedback lands differently with different people

One agent wants the detail and time to think it through. Another wants the headline and a quick plan. When the delivery doesn't suit the person, the effort put into the conversation doesn't always turn into change. For more on this, see how do you motivate call centre agents?

What would a sustainable coaching rhythm look like?

Whatever tools a centre uses, a few practical choices make a regular rhythm easier to sustain.

Agree a rhythm for each team. Decide how often each agent should have a coaching conversation, and give new starters and agents who want more support more frequent conversations.

Keep each conversation short and focused. Fifteen minutes is enough when it is regular. Open with what went well, look at one moment worth revisiting, agree one action and check on it next time.

Plan coaching into the workforce schedule. Treat it as planned time away from the phones, so it holds when other work competes for the hours.

Use the scoring you already do. If your centre already scores every agent monthly to meet ISO 18295 or COPC, those scored calls are ready-made material for coaching. The assessment is done. The conversation is the part that needs time.

Hold a lighter check-in when there is no new call. Some weeks there is no call worth discussing. A short conversation about goals, colleagues and progress on the agreed action still counts, and it covers the topics Gallup finds people value most.

How do you measure coaching frequency?

Track two numbers for each team: the share of agents who had a coaching conversation in the past seven days, and the share who had one in the past 30 days. The first shows how close the team is to the weekly rhythm Gallup describes. The second shows whether it is meeting the monthly floor the standards set. Count conversations held, rather than calls scored, because the two can differ.

For ways to structure the conversation itself, see our reviews of the GROW model and The One Minute Manager.

Where Sariio one2one fits

Sariio is the work preference layer. It sits between the call evidence a centre already has, the recording, the analytics and the QA scorecard, and the coaching conversation. one2one is the part of Sariio that prepares that conversation, at the cadence you set.

The preparation is done before the team leader starts

one2one reads the call and scores it against the campaign's own QA criteria, one criterion at a time. It then prepares a short note for the team leader that quotes what went well, names the moment worth revisiting alongside the standard behind it, and proposes one action for the agent's next call. The team leader reads it, changes what needs changing and endorses it. The agent sees nothing until then, and the conversation remains the team leader's.

In the example above, most of the 55 minutes went on finding, scoring and planning. With that done, the team leader's time goes on the conversation. You can see a worked example in Alex's one2one.

Each note is shaped to how the agent receives feedback

Each agent and team leader answers about twelve minutes of questions online to build their myMAP, a map of how they prefer to work. one2one uses the agent's myMAP to shape how the point is put: whether to send the moment ahead, whether to open with a question, what to avoid. The campaign standard is the same for everyone, and no preference changes a score.

That is the kind of tailoring Gartner's research describes, applied to every note rather than left to whoever has time to think about it. Read more about myMAPs.

Each conversation builds on the last

Each one2one ends with one action the agent agrees to try. one2one records where the agent started on that criterion, so the next scored call shows whether it has moved. Between conversations, the agent can carry on with Sariio Coach, and the team leader can ask Ask Sariio how best to open the next one.

It works from the scoring you already do

For a centre already working to ISO 18295 or COPC, one2one turns the monthly assessment you already carry out into material for more regular conversations. It works alongside your recording, analytics and QA tools, and doesn't replace any of them. Coaching time on the floor still needs protecting in the workforce plan. What changes is how far each coaching hour goes.

For a fuller comparison with other approaches, see how Sariio compares. For how we handle call recordings and agent data, see Trust and how Sariio uses AI.

What does it take to start?

The simplest way to see whether a more regular rhythm works for your centre is to try it on one campaign.

  1. Note where you are. Before you begin, record the share of agents who had a coaching conversation in the past seven days and in the past 30 days.
  2. Start with a small group. Begin with three to five agents and their team leader from one campaign. Each answers about twelve minutes of questions online, and we prepare one2one notes from their customer conversations against your campaign's own scoring criteria.
  3. Compare after a few months. Look at both numbers again, alongside the quality scores on the criteria you coached.
  4. Agree the call data first. For outsourced campaigns the recordings usually belong to the brand, so permissions are agreed with the brand before any calls are shared.

The first coaching comparison carries no charge and no obligation to continue. If you go ahead, the first campaign has a fixed scope and fee set out in a written proposal. After that, Sariio costs no more than £15 a person a month, with no contract (pricing).

To talk through your campaign, book a conversation with David, or read more about working with Sariio.

Sources