The Case for Auditing Your Homepage Every Quarter

Nobody breaks their homepage on purpose. It breaks the way a garden goes wild: not through any single act, but through the accumulation of small things nobody was watching. A marketing hire adds a banner for a webinar that ended in March. A plugin update changes how images load. Someone edits the headline in the CMS and the mobile layout quietly starts clipping the last word. Each change was reasonable. The sum is a page that no longer does its job.

The strange thing is that the people who own the page are the least likely to notice. You look at your own homepage the way you look at your own face: constantly, and therefore not at all. Familiarity doesn’t just breed contempt, it breeds blindness. You stop seeing the page and start seeing your memory of the page. A first-time visitor has no such memory. They see the expired banner, the clipped headline, the button that takes two seconds to become clickable. And they leave without filing a bug report.

This is why I’ve come to believe audits should be recurring, not occasional. A one-time audit answers the question “how is my site today?” But that’s the wrong question, because “today” is a moving target. The right question is “how is my site drifting?” and you can only answer that by measuring at intervals and comparing. It’s the difference between a photograph and a time-lapse. The photograph tells you where you are. The time-lapse tells you where you’re headed, which is almost always the more important thing to know.

Why quarterly, specifically? Mostly because it matches the rhythm at which sites actually change. Weekly is too often for most small businesses; the page hasn’t moved, and you’re paying attention to noise. Annually is too rare; a whole year of drift is enough for a redesign’s worth of decay, and by then the individual causes are impossible to untangle. A quarter is about the interval at which a few real changes have accumulated but each is still recent enough to identify and reverse. It’s also, not coincidentally, the interval at which most businesses already review everything else. Your homepage is a salesperson who talks to more prospects than your entire team combined. It seems odd to give every human on the payroll a quarterly review and never give one to the page.

There’s a second reason for the schedule, and it has nothing to do with your own changes: the ground moves under you. Browsers update. The devices your visitors carry change size and speed. Search engines revise what they reward. Accessibility expectations that were fringe five years ago are now table stakes, and in many jurisdictions, legal exposure. A page that was fine when you shipped it can become deficient while sitting perfectly still, the way a house up to code in 1995 isn’t up to code now. No commit in your repository will explain this kind of regression. Only re-inspection catches it.

When I built the recurring audit into GazeSite, the design followed from this reasoning. Each run looks at the page the way a stranger would: it loads the URL fresh, captures the rendered page and the underlying HTML, on desktop and on a phone-sized viewport, and then reviews it from six angles, from accessibility to conversion to the technical plumbing. The findings come back with severity and a prescribed fix. But the part I care most about is the interval. The product is not really the report. The product is the habit. A report you get once is trivia; a report you get on a schedule is a feedback loop.

The objection I hear is: I’d rather just be careful. Careful people don’t need scheduled inspections. But this misunderstands what kind of problem drift is. Being careful protects you against errors you can anticipate. Drift is made of errors nobody anticipated, interacting in ways nobody designed. That’s why even careful industries, aviation, medicine, elevators, rely on scheduled inspection rather than the diligence of practitioners. Not because the practitioners are sloppy, but because the failure mode is invisible from the inside.

The other objection is cost, and here I’ll just make the comparison plain. The cost of a quarterly audit, whether you do it with a tool or by paying a person for an hour, is small and fixed. The cost of an unnoticed problem is unbounded and compounding. Every day your signup button is broken on mobile, you lose everyone who would have signed up on mobile that day, and you never find out. There’s no error message for revenue that didn’t arrive. The audit is how you buy an error message.

So put it on the calendar. First Monday of the quarter, look at your homepage as if you’d never seen it, or better, have something that has never seen it look for you. Read the findings. Fix the worst three. Do it again in three months. It’s not glamorous. Neither is flossing. But the whole point of maintenance is that it’s boring right up until the moment you skipped it.