What Is the Control/Original?

The control is the page with your current design ,the version that already exists, before you change anything.

Every experiment needs one. The control is the baseline FigPii measures your new designs against, so you can tell whether a change actually improved your conversion rate or simply performed the same as what you already had.

An example

Suppose your homepage doesn't display any customer testimonials. You've heard that adding testimonials can lift conversion rate, but you don't know whether that holds true for your site.

So you set up an A/B experiment:

The control is your current homepage, with no testimonials.
The variation is a new version with testimonials added, which you create in the FigPii visual editor.

A variation is sometimes also called a treatment.

When you launch the experiment, FigPii divides your visitors between the control and the variation. Some visitors see your original homepage, some see the new one , and neither group knows an experiment is running.

What FigPii reports

FigPii gives you a report showing, for both the control and each variation:

How many visitors saw it
How many of those visitors converted
The resulting conversion rate, and how each variation compares to the control

It uses those visitor and conversion numbers to determine whether the experiment has produced a winner. See What Is A Conversion For The Testing Engine? and What Does Minimum Number Of Conversions Before Declaring A Winner Mean?

Why the control keeps receiving traffic

A common question is whether the control still gets visitors once an experiment is running. It does — and it has to.

Conversion rates move for all sorts of reasons that have nothing to do with your test: seasonality, a marketing campaign, a change in traffic mix, a slow week. Running the control at the same time as the variations means all versions experience those same conditions, so any difference you measure comes from the design change rather than from what else was happening that week.

Comparing this month's variation against last month's numbers would not tell you the same thing.