Dynatrace provides a built-in forecast based on your consumption patterns, and you can enhance it with business context to build a budget-adjusted projection.
FinOps practitioners, engineers, and platform team leads who want to understand the Dynatrace built-in forecast, decide when to trust it, and enhance it with business context they already know.
In this tutorial, you'll learn how to:
Once you have at least 15 days of consumption history, Account Management provides automated forecasts. To see these, go to Account Management > Subscription > Overview.
The built-in forecast becomes available once your account has 15 or more days of cost data. For newer accounts or after major environment changes, use the DQL run-rate tutorials instead. For more information, see Forecast run rate.
Here's what's included in an automated forecast:
| Component | Description | Use case |
|---|---|---|
Median forecast | Most likely end-of-period consumption. | Primary planning number. |
Upper bound | Higher-end projection accounting for variability. | Conservative / worst-case planning. |
Lower bound | Lower-end projection. | Best-case scenario. |
Forecast event | Alert when forecast exceeds commitment. | Early warning system. |
The built-in forecast is reliable for stable, mature environments with consistent usage patterns. Treat it with caution when:
In these cases, layer your own business context on top, or use DQL-based run-rate queries for a more grounded projection.
Dynatrace forecasts are consumption-based. They don't know about your roadmap. You add value by layering in what you know.
Here's how to use your business context to enhance the automated forecasts:
Dynatrace median forecast (baseline)+ Planned workload additions- Planned retirements / decommissions+ Anticipated organic growth± Seasonal adjustments─────────────────────────────────────= Business-adjusted forecast
And here's that calculation with some example figures:
| Component | Value |
|---|---|
Dynatrace median forecast (end of period) | $480,000 |
+ New application launching Q3 (estimated) | +$40,000 |
− Legacy system retirement Q2 (estimated) | −$25,000 |
+ Organic growth adjustment (10%) | +$48,000 |
Business-adjusted forecast | $543,000 |
Document your adjustments, because serve as a record for forecast accuracy reviews at the end of the period.
This section is relevant if you have a Proof of Concept (POC). For everyone else, you can skip to the next step.
POC consumption often misleads planning. A single, lightly configured environment with minimal retention and everything enabled for exploration is not representative of your production footprint. Before you treat your POC forecast as a baseline, account for these factors:
| Factor | Typical in POC | Typical in production | Planning adjustment |
|---|---|---|---|
Environments | 1 | 3+ (dev, staging, prod) | Multiply by number of environments |
Host / pod scale | Small sample | Full fleet | Scale up proportionally |
Retention settings | Default (short) | Compliance-driven (longer) | Add retention costs for each capability |
Features enabled | Everything on for trials | Right-sized per environment | Often lower per unit than POC suggests |
Cost allocation | Not configured | Tags applied | No cost impact, but needed for showback |
Here are tips for how to successfully transition from a POC to production environment.
For a per-host cost estimate based on your existing fleet, use the DQL approach in Forecast costs for new resources.
You might notice that your actual costs consistently vary from the automated forecasts. Here's how to identify when you need to act:
| Situation | Signal | Action |
|---|---|---|
Slightly under (<10%) | Normal variance | Monitor; no action needed. |
Significantly under (<20%) | Potential under-use | Review whether value is being left on the table. |
Consistently under | Over-committed | Consider commitment reduction at renewal. |
Trending over (10–20%) | On-demand usage likely | Initiate optimization review; alert stakeholders. |
Significantly over (>20%) | Budget risk | Escalate, implement optimization measures, consider commitment adjustment. |
You can now use the built-in Account Management forecast as part of your DPS planning. Specifically, you can: