Case: a 30-60-90 plan for the Support team

A 30-60-90 plan is a new lead's commitment for the first three months, split into three 30-day blocks. On one page it answers four questions: what you stop doing yourself, what you hand over and to whom, what you watch to know it's working, and what you need your manager to approve.

The blocks follow the calendar, not a template. At Klasna days 1–30 are February (listening, the team agreement, your own week), days 31–60 are the March peak with two new seasonal hires, and days 61–90 are April, closing with a review with Iryna. Check-ins with her fall at day 30 and day 90.

A good plan is short enough to read before a meeting, 8 to 12 lines, and every line names a person, a number or a date. “Improve team communication” isn't a line; “Agree who decides refund exceptions by 14 February” is.

Write it so the team could read it too, because in a remote team half of them will.

The first line of a new lead's plan should be about the one thing the lead fully controls: their own week. A plan that changes everyone except the lead rarely survives a busy Monday, because the lead's habits pull the work straight back.

Use the gap between the calendar audit and a target week from Your output is now the team's output. At Klasna, week 4 (25 January) already moved 8 hours: own tickets 23 → 15, enabling 11 → 19. Escalations stayed at 9. The 40-hour target: own tickets 8, escalations 5, enabling 22, admin 5.

So the plan's first line reads: own tickets 15 − 8 = 7 more hours a week to hand over by day 30, escalations 9 − 5 = 4 fewer, enabling 22 − 19 = 3 more, and the week back from 47 hours to 40.

Each of those hours has to land on someone, which is why the next line of the plan is about who takes what.

Handing over work means handing over decisions, and a decision needs limits: who decides, within which rule, and how you'll know what was decided.

At Klasna, Olha's numbers show what the team lost in January: 14 hard tickets in December, 6 in January, 4 last week. She asks for something specific: if refund exceptions come to her, she decides them rather than preparing them for your sign-off. That's a fair request, and the plan can grant it with a clear line: Olha decides refund exceptions within the refund policy and reports them weekly; anything outside the policy is Iryna's call.

The limit is what makes the handover safe. The policy sets the edge, the weekly report keeps you informed without approving each case, and the escalation route above the policy is named in advance.

As Iryna puts it, she'd rather approve one decision rule than a hundred tickets. A lead who keeps approving every case hasn't handed anything over.

The steps of a 90-day plan depend on each other, so their order matters as much as their content.

Listen first. The one-to-ones tell you why people hand tickets up, and why Bohdan's first replies take over half an hour. Without them, every later step rests on guesses.

Then agree. A written team agreement on what each coordinator decides alone and what comes to the lead grows out of what you heard.

Then step back. Only once everyone knows who picks up what can you safely cut your own tickets to 8 hours a week, with hand-ups checked weekly. Cutting first leaves tickets with nobody.

All of that belongs in February, days 1–30. The seasonal hires arrive in March (days 31–60) and need an onboarding plan that relies on the agreement already working. The review with Iryna closes the 90 days in April.

A plan in the wrong order fails even when every step is sensible on its own.

Team numbers in a plan should point to questions, not settle them. The most common trap is a rate built on a small base.

At Klasna, over December–January Taras closed 726 tickets, a third of the team's 2,145, and Maryna 453. But 9.50% of Taras's tickets reopen against Maryna's 1.77%, and her positive survey share is 94.78% against his 78.24%. Tempting to declare Maryna the quality star.

Look at the base. Satisfaction counts survey answers, not tickets: Maryna has 134, Taras 216. With 134 answers, a single answer moves her rate by 1 / 134 ≈ 0.75 percentage points. A few answers either way would shift the picture. It's a strong signal worth asking about, not a verdict to build a role on.

The same caution applies to Bohdan: he replies slowest, 36.13 minutes on average, yet only 3.72% of his tickets reopen. That's a question for his one-to-one, not a conclusion for the plan.

When a manager asks for one team rate, the honest number is pooled: all reopened tickets over all tickets closed. The tempting shortcut, the plain average of personal rates, weighs everyone equally whatever their volume.

At Klasna, Iryna wants one reopen rate for her report. Reopened: Olha 23 + Taras 69 + Maryna 8 + Bohdan 15 = 115. Closed: 563 + 726 + 453 + 403 = 2,145. Pooled: 115 / 2,145 ≈ 5.36%. The mean of the four personal rates, (4.09 + 9.50 + 1.77 + 3.72) / 4, is 4.77%: lower, because it counts Bohdan's 403 tickets as much as Taras's 726, and Taras has the highest rate.

The pooled rate is what families actually experienced: about one ticket in nineteen came back.

Report the team rate upward. Keep personal rates for one-to-ones, where they start a conversation about how someone works, and never put them in a league table.

A plan written late on a Sunday usually carries the lead's worst instincts. One test catches most of them: read it as if it had leaked into the team chat. Would Taras or Maryna trust a lead who wrote it?

Four kinds of line fail that test.

• A ranking by one metric: “Monthly ranking by tickets closed; Taras sets the standard” rewards speed over resolution and ignores his 9.50% reopen rate.
• Surveillance: “Install activity tracking to see who is really working” measures presence, not the work, and costs trust you won't get back.
• A role from one number: making Maryna quality lead from 94.78% on 134 survey answers, before even asking her.
• A wrong team rate: the plain average (4.77%) instead of the pooled 5.36%.

Lines that pass are plain: cut your own tickets to about 8 hours a week by day 30; ask each person what slows their work and what they'd like to own.

As Olha says for the team: people can live with being measured, not with a ranking by one number nobody explained.

The last block of the plan lists what your manager has to approve. Two kinds of item belong there: anything outside your authority, and anything that costs the team or the business something.

At Klasna that means: decision rights above your level, such as refund exceptions outside the policy, which stay Iryna's call; the expected cost of the change, since first replies may slow for about two weeks while hand-ups move to the rotation, and Iryna should hear that from you before she hears it from a family; and rota hours for onboarding Dmytro and Sofiia in March.

What doesn't belong: things inside your own authority, like how you run one-to-ones or move your own hours. Asking approval for those hands your job back up.

Iryna's own words make a good template: what you'll do, what it costs, what you need from me.

How it fits together: your own week, decision rights with limits, the right order, honest numbers and an approval block make one page Iryna can say yes to. The next module, February, starts the one-to-ones and the feedback conversations this plan depends on.