There is no single good conversion rate. The phrase hides three different measurements wearing the same name: how many site visits end in a purchase, how many landing page visits end in an opt in, and how many ad clicks end in a tracked action. This quarter, ecommerce sites averaged 1.4 percent site to purchase. Landing pages averaged 6.6 percent visit to opt in. Both numbers are true. Neither tells you anything about the other.
The reason both can be true comes down to the denominator, the number you're dividing by. A site to purchase rate divides by every visitor who ever landed on the site, most of whom were never going to buy anything today. A landing page rate divides by a narrower, more qualified group: people who already clicked an ad or a link built around one offer. Google Ads goes narrower still. It defines conversion rate against ad interactions, not site visits at all, so a client quoting 5 percent from their ads dashboard isn't describing the same thing as a 5 percent site wide rate. Most conversion rate anxiety comes down to comparing two figures that were never measuring the same funnel.
The published ranges bear this out. Landing page conversion runs from 3.8 percent for SaaS to 12.3 percent for events and entertainment, most industries falling somewhere between. Ecommerce sectors cluster far lower, close to 1.4 to 2.4 percent, because that number has to survive the whole path from browsing to paying. Traffic source moves the figure on its own. Email traffic on the same page converts several times higher than paid search or display, with nothing on the page itself touched. Price, how complex the sale is, and how much trust a visitor needs before acting: all of it shifts the number before design even enters the conversation.
The only conversion rate worth chasing is your own: measured against your own history, on the same definition, over time. A rate that drops below your baseline is diagnosing something, not decorating a report. Sometimes it's a traffic mix that changed. Sometimes a campaign pulled in the wrong audience, or a funnel stage got redefined without anyone noticing. A marketer can check all three alone, no design work required. What's harder to see from inside the business is whether the page still matches the person arriving on it, once those easier explanations are ruled out. That's usually where a studio earns its fee, not a redesign nobody asked for.
So before touching the design, know which number you're actually holding: site wide, landing page, or ad platform. They aren't interchangeable. Compare it against your own history, not an industry chart built from someone else's traffic and someone else's price point. If the honest read is still confusing after that, it's worth a second pair of eyes before a rebuild.