Forge Mark

Know if the change paid for itself

Sales went up after you shipped. Was it the change, or the season? Forge Mark compares 28 days before and after against the rest of your store.

Each change you ship, measured over 28 days before and after against the rest of the store.

Forge Mark's Fixes screen on a made-up business, with sample data.
Forge Mark itself on a made-up business, not a video. Click anything; nothing you do here is stored. More on the demo page.

Every shipped change gets a verdict

A figure with a range, or a plain answer that it can't tell.

Each change is measured over the 28 days before and after, against the pages it didn't touch. A good month for everyone isn't credited to one change.

The result sits beside the fix that caused it.

Explore impact

ChangeDelivery dates above the add to cart button

Orders a day, the 14 products it touched

Change shipped

18 Aug15 Sept12 Oct

These productsAfter the changeThe rest of the store, scaled to themWhat would have happened, with its range

+18%

Probably 18% more orders of these productsSomewhere between +9% and +27%, compared with how the rest of the store did

28 days before, 28 days after

Illustration on sample data.

Prove an ad with a holdout test

Ads off in some regions, on in others.

A wizard sets it up and the result comes back in plain words with a range. The gap in sales is what the ads added.

When a store is too small for a test to tell, it says so before you start.

Explore measurement
  1. 01Pick regions
  2. 02Ads off in test
  3. 03Weekly sales
  4. 04The lift
  • ABon
  • IVon
  • Goff
  • EHon
  • NEon
  • LAon
  • LSoff
  • YOon
  • Mon
  • NGoff
  • Bon
  • NRon
  • CFoff
  • BSon
  • SOon
  • BNoff

Test: 5 regions, ads offControl: 11, ads on

Sales a week, per region

Control regionsTest regionsRange

+£4,400

a month, most likely

between +£2,300 and +£6,900 a month

The whole range is above zero

+£300

a month, most likely

between -£2,100 and +£2,700 a month

Too small to measure.

Illustration on sample data.

Ask buyers what the clicks missed

A podcast, a friend, TV.

A post-purchase survey asks where they heard of you. Results show by channel beside the tracked clicks, once at least 10 people have answered.

Survey is one of six lenses in Measurement.

Explore surveys

How did you hear about us?

responses: 214

Clicks saySurvey says

  • Instagram34% of tracked clicks22% of answers
  • Google38% of tracked clicks21% of answers
  • Email14% of tracked clicks6% of answers
  • A podcast1% of tracked clicks17% of answers
  • A friend2% of tracked clicks19% of answers
  • TV0% of tracked clicks9% of answers

45%of answers name a podcast, a friend or TV

3%of tracked clicks came from them

Not shown: something else, 6% of answers. No click, 11% of orders.

Illustration on sample data.
28
days before and after, for every shipped change
10
answers before survey results are shown
Range
on every result, and "too small to measure" when that's the answer
£0
a month

Frequently asked questions

Why 28 days?

Four whole weeks cover every day of the week four times, which smooths out a busy weekend or a payday.

What counts as a change?

A fix you apply in the app, a change you made yourself, or a job Forge finishes. Impact measures each one against the rest of the store.

Is a holdout test better than the before-and-after?

It's firmer, because it runs both at the same time. It also needs enough sales in each region, which is why Forge Mark offers both.

Does Forge Mark switch my ads off for a test?

No. The wizard tells you which regions to hold out and for how long. You switch the campaigns off and on yourself, and Forge Mark reads the result.

Measure your next change.

Connect your store in a few minutes and let the next fix prove itself.