# Understanding your forecasting data

**Understanding Forecasting: See 12 Months into Your Hybrid Future**\
(How gospace gives you 90–94 % accurate visibility of exactly who is coming in, when, and how much space you really need)

#### Why Forecasting Is the Foundation of All Three Value Pillars

Accurate forecasting is what makes the other two promises possible:

| Without accurate forecasting                      | With gospace forecasting (90–94 % accuracy)                                      |
| ------------------------------------------------- | -------------------------------------------------------------------------------- |
| You over-allocate desks by 40–60 % “just in case” | You know the exact number needed every day → safe consolidation                  |
| You guess which teams will be in together         | You maximise natural co-attendance and co-location                               |
| Real-estate decisions are based on gut feel       | You have board-ready data to close floors, renegotiate leases, or open new space |

Clients typically move from “We think Tuesdays are busy” to “We know exactly how many desks and meeting rooms we need every day for the next 12 months” within the first week

#### The Forecasting Dashboard – Your Single Source of Truth

Go to **Admin → Forecasting** (or click the crystal-ball icon in the main navigation)

| Setting                         | What it shows                                                                                       | Typical use-case                                                                    |
| ------------------------------- | --------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------- |
| **Date or date range**          | Single day → 12 months ahead                                                                        | “Show me every Wednesday from now until December 2026”                              |
| **Average vs. Peak view**       | Average = typical day; Peak = 95th percentile (what you actually need to plan for)                  | Average = day-to-day ops; Peak = lease & consolidation planning                     |
| **Granularity toggles**         | <p>• By Location / Building / Floor<br>• By Team (Level 3–8)<br>• By Individual (probability %)</p> | “How many desks does Engineering-L7 actually need on Thursdays in Q3 2026?”         |
| **Day-of-week heat-map**        | Colour-coded Mon–Fri grid showing forecasted headcount (darker = busier)                            | Instantly spot the new “Wednesday + Thursday” peak pattern                          |
| **Team demand table**           | Exact desk & room demand per team, per day of the week, for the selected period                     | Export this table straight to your real-estate committee                            |
| **Individual probability view** | Every employee’s % chance of being in on any given day (with historical vs. forecast comparison)    | “Sarah now has an 87 % chance of being in on Wednesdays – up from 34 % pre-gospace” |

#### What You Can Answer in <30 Seconds

| Question Leadership Always Asks          | Where to click                                                       | Real client example (GSK)                                           |
| ---------------------------------------- | -------------------------------------------------------------------- | ------------------------------------------------------------------- |
| How many people will be in next quarter? | Select date range → Peak view                                        | “1 842 average, 2 631 peak on Wednesdays”                           |
| Which teams drive the peaks?             | Toggle “By Team” + Peak view                                         | “Data Science + Trading are the two biggest drivers on Wed/Thu”     |
| Can we close Floor 8 from September?     | Run Consolidation Simulation using the forecast as the demand source | Saved £2.7 m (actual)                                               |
| What does Christmas week look like?      | Select 22–31 Dec → Average view                                      | “<18 % of normal headcount – perfect time for maintenance shutdown” |
| How is the new 4-day RTO policy landing? | Compare historical vs. forecast probability for individuals          | “Average individual probability up from 61 % → 84 % on in-days”     |

#### Accuracy & Continuous Improvement

| Forecast horizon | Typical accuracy (live client average) | How we keep improving it                           |
| ---------------- | -------------------------------------- | -------------------------------------------------- |
| Next 7 days      | 94–97 %                                | Calendar + intention signals + same-day badge data |
| 8–30 days        | 91–94 %                                | Historical patterns + meeting invites              |
| 31–90 days       | 89–92 %                                | Seasonality + policy changes                       |
| 91–365 days      | 84–90 %                                | Long-term trends + headcount planning data         |

The model retrains daily and automatically incorporates any new policy (e.g., “Wed–Thu core days”) the moment you set it.

#### Quick Actions from the Forecasting Dashboard

* Export → CSV / PDF / PowerPoint (board-ready in one click)
* Feed directly into a Consolidation Simulation
* Create a “Forecast vs. Actual” report (proves ROI to leadership)
* Share a read-only link with your real-estate partner or broker

#### Next Step for You

1. Open **Forecasting** right now
2. Select the next 12 months → Peak view → By Team
3. Screenshot the day-of-week heat-map and the top-10 teams table

That single screenshot is worth millions in lease negotiations.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.gospace.com/overview/trial-subscription-guides/understanding-your-forecasting-data.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.