Changes blur together
Fifty products edited in one month, and nobody can say which edit did anything.
Impact & attribution · Before/after impact
Every applied change gets its own before and after.
Sphen.ai compares 14 days before and 14 days after each change it writes to Shopify, and tells you whether it improved, worsened or did not move.
In short
What is before/after impact in Sphen.ai?
Before/after impact is the measurement Sphen.ai runs on every change it writes to Shopify. It compares 14 days before the change with 14 days after it, then sorts the change into Improved, Worsened, No measurable movement or Insufficient data. While data is still coming in, the change shows Measuring. Measurement only reads. It never changes your store.
Updated:
The problem
Without a baseline, every fix is an anecdote.
Success by feeling
A good week after a change proves little. A defined window before and after proves more.
No stop signal
If a type of change does not help, you want to know before you roll it out to the whole catalog.
How it works
How does before/after impact work?
-
01
Know it before you apply
Every recommendation says what happens when you apply: it writes only the listed fields and measures 14 days before and after.
Demo store · sample data -
02
Measuring, then a result
Applied changes show Measuring while data accrues — about two weeks. The result appears on the change in History.
Demo store · sample data -
03
Read the tally
Measured impact on Overview sorts changes into Improved, Worsened, No measurable movement or Insufficient data, with improvements confirmed at the top.
Demo store · sample data
In the app
Where do you see the measurement in the app?
On the recommendation, before you apply
Every recommendation states what happens when you apply it: it writes only the listed fields and measures 14 days before and after. You also see Before / after and the Evidence block with Current, Target and Signal, so you know what is being changed and why. Applying a draft is what starts the measurement window. Drafts you dismiss or never apply are not measured.
In History, while it is measuring
Under Recommendations → History, each applied change carries its status. Measuring means the change is live and Sphen.ai is waiting for enough data to judge it. That window is normal and runs about two weeks. The result then appears on the change. A Live on store badge verifies the value against your real storefront, and Differs on store tells you someone edited the field in Shopify afterwards.
On Overview, as a tally
Once your first applied change has been measured, a Measured impact card joins the Overview page. It sorts measured changes into Improved, Worsened, No measurable movement or Insufficient data, with confirmed improvements at the top. Expect a mix. If a type of change did not help, you can revert it from History before you roll the same kind of edit out to more products.
You stay in control
What stays under your control?
- Measurement only reads. It never changes your store.
- Insufficient data is reported as such, not rounded up to a win.
- A change that did not help can be reverted from History with one click.
Plans
Which plans include it?
Included in every plan, with no separate quota.
How long does measuring take?
About two weeks after a change goes live. The status Measuring during that window is normal: the change is already live on your store, and Sphen.ai is waiting for enough data to judge it. The result then appears on the change under Recommendations → History.
Will every change show an improvement?
No. Expect a mix of improved, worsened and no measurable movement. The useful number is improved minus worsened over a quarter, not the result of a single change. Changes without enough data are reported as Insufficient data and are not counted as a win.
What if someone edits the field in Shopify afterwards?
The change is marked Differs on store, so you know the measured version is no longer the live one. If you then revert it from History, Sphen.ai warns you that the store content changed, because reverting would overwrite the newer edit made in Shopify.
Does measuring change anything in my store?
No. Measurement only reads. The only write happens when you apply a change, and it covers just the fields listed on the recommendation. If a measured change did not help, you decide whether to keep it or revert it under Recommendations → History.
What does Insufficient data mean?
It means there was not enough data in the window before and after the change to judge it. Sphen.ai reports that as its own result and does not round it up to an improvement. The change itself stays applied on your store unless you revert it.
Where do I find the result for a single change?
Open Recommendations → History. Each applied change shows its status there, first Measuring and then the result. The Measured impact card on Overview gives you the tally across all measured changes. It appears once your first applied change has been measured.
See it on your own catalog.
Install Sphen.ai, let it read your store, and decide exactly what goes live.