Measuring After You Ship

Your feature is live. Now you want to know: did it work? Post-release metrics capture the “after” — letting VEKTIS compare what the metric looks like now against what it looked like before.

Just like with baselines, how these measurements arrive depends on how the dev item is measured. By default, VEKTIS collects them for you: if the dev item is set to “VEKTIS measures it for me,” post-release measurements flow in from your product’s analytics automatically — nothing to type in. To enable this, add the VEKTIS tracker to your product (see the Quickstart, whose Use AI option can wire it up for you).

Entering values by hand is optional — for dev items you measure yourself (“I’ll enter results myself”). You pull the metric from whatever source you normally use — analytics dashboard, observability tool, internal reports — and enter it into VEKTIS, using the same source and method you used for the baseline so the comparison is apples-to-apples.

Prerequisites

Before you can add post-release metrics, you need:

1. A release date — The dev item must be marked as released (status: Measuring or Complete)
2. Baseline data (recommended) — At least one baseline metric recorded (2+ recommended for meaningful results). If you don’t have baseline data, VEKTIS can establish a baseline automatically from your first few post-release measurements.

Adding a post-release metric by hand

These steps are for dev items you measure yourself. If VEKTIS measures the dev item automatically, post-release measurements arrive on their own and you can skip this.

1. Click on your dev item from the dashboard to open its detail page — you’ll find the Post-Release Metrics section below the baseline data
2. Click “Add Post-Release Metric”
3. Fill in the fields:
   • Date — When this measurement was taken. Must be on or after your release date, and can’t be in the future.
   • Value type — Must be compatible with your baseline type (see Metric Value Types)
   • Value — The actual measurement
   • Notes — Optional context (up to 500 characters)
4. Save — VEKTIS immediately calculates the delta, impact score, and interpretation

What happens when you save

Every time you add or update a post-release metric, VEKTIS automatically calculates:

• Delta — The difference between this measurement and your baseline average
• Impact score — A 0–100 score indicating how significant the change is, from Signal Detected (normal range) through Confirmed Impact to Significant Impact (strong evidence)
• Direction check — Whether the change is moving toward or away from your target
• Interpretation — A plain-language summary of what the result means

These calculations update instantly — no waiting or manual refresh needed.

Date rules

Post-release dates follow a simple logic:

• Must be on or after your release date — you can’t record a “post-release” measurement from before you shipped
• Can’t be in the future — measurements should reflect actual observed data
• One per date — Each date can only have one post-release metric. Edit the existing entry if you need to correct a value.

Type compatibility

Post-release metrics must use the same value type as your baseline (or a compatible duration type). This ensures the comparison is valid.

• If your baseline uses percentage, post-release must also use percentage
• If your baseline uses any duration type (seconds, minutes, hours), post-release can use any duration type — VEKTIS converts automatically

Note

The value type is determined by the first metric you add to a dev item — whether baseline or post-release. You can see the current type in the baseline metrics section of your dev item.

Building confidence over time

A single post-release measurement gives you a snapshot. Multiple measurements reveal a trend.

Post-release data points	What you can learn
1	Initial signal — interesting but not conclusive
2–3	Pattern emerging — if results are consistent, confidence grows
4+	Clear picture — you can make decisions with confidence

Tip

Space your measurements out. Weekly data points over a month tell you more than daily data points over a week, because they capture more natural variation in the metric.

Starting without baseline data

Sometimes you don’t have pre-release data — maybe the feature is entirely new, or you started measuring after it shipped. VEKTIS handles this with a burn-in process that establishes a baseline from your initial post-release measurements.

Here’s how it works:

1. Collecting measurements — You add post-release data points as usual. VEKTIS shows a progress indicator: “Establishing baseline from post-release data” with a count of how many measurements you’ve added so far.
2. Baseline established — Once you’ve added 3 measurements, VEKTIS uses those first data points as the baseline. This is equivalent to having collected baseline data before release.
3. Scoring begins — From the 4th measurement onward, VEKTIS calculates impact scores by comparing new data points against the established baseline — just like it would with pre-release baselines.

Tip

While the burn-in process works well, collecting baseline data before you ship is always better when possible. Pre-release baselines give you a true “before” picture, while burn-in baselines represent the metric shortly after release — which may already include the feature’s effect.

Note

Your first 3 measurements become the baseline. After that point, impact scores appear on all subsequent measurements — and you’ll see the same impact zones (Signal Detected, Confirmed Impact, Significant Impact) as with traditional baselines.

Editing and deleting post-release metrics

You can edit or delete post-release metrics while the dev item is in Measuring or Complete status.

• Editing — Change the date, value, value type, or notes. VEKTIS recalculates the impact automatically.
• Deleting — Removes the data point. Remaining metrics keep their individual calculations.

Common questions

Can I add post-release metrics to a dev item that’s “Complete”? Yes. Moving to Complete doesn’t lock out new measurements — it just signals that you consider the measurement period finished. You can still add data if needed.

What if my first post-release result shows “Signal Detected”? Don’t worry — that means the change is within the normal range so far. Keep measuring. A real impact often takes a few data points to become statistically clear, especially if the metric naturally fluctuates.

What if the impact direction is wrong (red indicator)? A red indicator means the metric moved opposite to your target. This is worth investigating, but one data point isn’t conclusive. If you see a consistent red pattern across multiple measurements, the feature may have had an unintended effect.