Glossary · AI systems & marketing
A/B test
Showing two versions of a page or message to see which one performs better.
What it means
An A/B test splits traffic between two variants that differ in one respect and compares a measured outcome, such as leads per visitor. It replaces opinion with evidence, within the limits of how much traffic a page receives.
Why it matters on the South Shore
ICS runs tests on client sites where traffic supports them and otherwise applies what has already been learned across South Shore trades and practices. The scoreboard reports what changed and what it did; opinions do not appear in it.
Straight answers
- Can a small business run meaningful A/B tests?
- On a page with a few hundred visitors a month, tests on big changes can read in a quarter; tests on button colors will never reach significance. Test the headline, the offer, the form length, and the proof placement. Skip the trivia.
- What should I test first?
- The first screen of your highest-traffic page: the headline, the call to action, and whether a price or a quote tool appears above the fold. Those move conversion by multiples; everything else moves it by percentages.
Related terms
-
Conversion rate
The percentage of visitors, leads, or quotes that take the next step.
-
Landing page
A single-purpose page built to turn a visitor into a lead for one specific service or offer.
-
Dashboard reporting
A live view of leads, response times, conversion, and booked work by source, reviewed monthly.
Where this term shows up
Terms that point here: Click-through rate (CTR) · Heatmap analytics