Use case

Support ticket forecasting for staffing and backlog planning

Forecast daily ticket volume from a CSV, read the weekday pattern and the 80% range, and plan agent coverage without overstaffing the average day.

Run the sample Put this in production

Support staffing is a queueing problem. The average day is not the day that breaks the SLA; the busy tail is. A forecast with a range lets a support lead staff to the median and keep a explicit plan for the high band.

The sample is daily created tickets with a Monday peak. If your queue is dominated by a few large customers, forecast those accounts separately; a single aggregate series hides the concentration that actually drives the workload.

The sample series

The studio below is preloaded with this daily tickets created series and a 14-step horizon. It is an illustrative example, not a real company’s data. Replace it with your own CSV to run the same workflow.

Show the CSV

date,tickets
2026-02-01,34
2026-02-02,52
2026-02-03,48
2026-02-04,45
2026-02-05,44
2026-02-06,41
2026-02-07,22
2026-02-08,24
2026-02-09,55
2026-02-10,50
2026-02-11,47
2026-02-12,46
2026-02-13,43
2026-02-14,23
2026-02-15,25
2026-02-16,58
2026-02-17,53
2026-02-18,49
2026-02-19,48
2026-02-20,45
2026-02-21,24
2026-02-22,26
2026-02-23,60
2026-02-24,55
2026-02-25,51
2026-02-26,50
2026-02-27,47
2026-02-28,25

CSV forecast studio

Forecast your support ticket forecasting series.

This studio is preloaded with the daily tickets created example above. Replace it with a CSV of your own history, choose a horizon, and run the forecast. The public demo is free while the Hugging Face Space is available; the API is the paid product.

1 · Paste or upload a CSV2 · Pick a horizon3 · Download the 21 quantiles

Steps

1 credit per API call · the web demo is free while the public Space is available

What to watch in the output

Created tickets, not closed tickets

Closed-ticket volume measures your team’s throughput, not incoming demand. Forecast the created queue; then compare it with capacity to see the backlog.

Plan the high band for SLA risk

If the high band exceeds the team’s daily capacity for several days in a row, the backlog will grow even if the median is manageable.

Keep channel definitions stable

A new self-serve help center can cut email volume without changing customer demand. A definition change is a structural break; note it and consider forecasting the post-change period only.

Common mistakes

—Mixing created and solved counts in one series. The two have different dynamics and different owners.
—Ignoring release days. A product release can create a predictable ticket spike; add it as a planned scenario rather than letting it look like noise.
—Forecasting at a weekly grain when staffing is daily. The aggregation hides the Monday peak that drives the schedule.

How to put it to work

1. Start with the sample. Run the preloaded series once so you see the output shape before you debug your own CSV.
2. Replace it with your history. Keep one timestamp column and one or more numeric columns. Mark outages and stockouts as empty cells, not zeros.
3. Backtest before you plan. Hold out the last few periods, compare the forecast with what happened, and check whether the actuals land inside the 80% band.
4. Move it into production. Use the REST API for a nightly job or an internal tool. One credit per successful call; failed calls are refunded.

FAQ

Can I forecast ticket volume by category?

Yes, up to eight related categories in one file. For independent categories, separate files give a cleaner per-category error trail.

Does the model account for response time?

No. It forecasts the arrival series you provide. Response time depends on staffing, queue discipline and ticket complexity; model it separately.

How much history should I use?

At least six to eight weeks of daily data to capture the weekday pattern. If you have a seasonal business, use a full year when available.

Need this as an API?

The same forecast is available as a metered REST endpoint with JSON and a downloadable quantile CSV.

See the API