A head-of-sales dashboard on amoCRM: counting plan and actuals so no one argues with the numbers

In short. A head of sales does not need yet another amoCRM report but an honest ranking: who is meeting the plan given what day of the month it is. To get there the plan is stored as a daily rate, actuals are counted from closed deals, and a row's colour reflects pace rather than percentage. Below is how we built such a dashboard for a client in 2024–2025 and which decisions paid off.

Where the numbers come from

Every hour the service pulls users, groups and roles from amoCRM, as well as each manager's deals. Actuals are the sum of won deals closed in the period; leads are deals created in the period; conversion is the ratio between them. Blocked employees are detected and kept out of the ranking.

The plan as a daily rate

A monthly plan is set for a group or a person and stored as a rate per day. The plan for any period is then the sum of the daily rates, even when the period spans two months with different plans. A group plan is split equally between the selected managers.

One subtlety: the rate is rounded to whole tenge, so ₸1,000,000 over 31 days becomes ₸999,998 for the month. The interface should account for it so the monthly total matches what the manager entered.

A traffic light by pace

A manager at 40% of plan on day 12 of a 30-day month is ahead. So a row's colour is based on the expected percentage: actuals divided by the share of days elapsed. Green means pace at or above plan, yellow means at least half, red means below half.

Where the numbers mislead

• Different cohorts. Sales are counted by close date, leads by creation date. Conversion compares different sets of deals and over a short period can exceed 100% — explain this upfront, or people stop trusting the dashboard.
• API limits. amoCRM allows no more than 7 requests per second per integration. Users are refreshed in small batches with pauses, recently refreshed ones are skipped, and long periods are computed in the background with a progress bar.
• Double checking. Deals are filtered in the API request and again locally — by the “closed won” status and by dates — so deals from the edge of the neighbouring period do not slip into the report.

What it gives the sales team

One number for everyone: how much has been sold, how much is left to plan and who is behind given the day of the month. Stand-ups are about deals, not about whose spreadsheet is right. We use a similar approach to amoCRM data in our lead-source report.