Setting a conversion goal before you redesign anything

A conversion goal before redesign is the single measurable action you want a visitor to take — a purchase, a form submission, a booking. Setting it first gives you a baseline to beat. Without that number, you cannot prove the new site worked. Everything else follows from it.

Key Takeaways

• A conversion goal is one measurable visitor action, not a wish list of "better engagement" or "more modern feel."
• You need a baseline before redesign; redesigning without one means you can never prove the money was well spent.
• Macro goals (sales, leads) matter more than micro goals (page views, time on site) when you decide whether the redesign paid off.
• Set the goal with the person who owns the revenue number, not just the designer or the marketing team.
• Track the same goal before, during and after launch using the same tool and the same definition, or your comparison is meaningless.
• A redesign that makes the site prettier but drops the conversion rate by 15% is a failed redesign, no matter how good it looks.
• Most failed measurement happens because nobody wrote down what "conversion" meant in week one.

What does "conversion goal before redesign" actually mean?

A conversion goal is a visitor action you can count and tie to revenue or intent. It is not a mood — "the site should feel more premium" is not a goal. It is a countable event: a completed checkout, a quote request, a phone call from the contact page, a booked appointment. Setting it before redesign means you record how often that event happens today, on the current site, before anyone touches a wireframe. That number becomes the yardstick for every design decision that follows.

The phrase matters because the sequence is non-negotiable. If you define the goal after the redesign is done, you are not measuring the redesign — you are rationalising it. You will pick whichever metric happens to look good. We have watched teams do exactly that: launch a new site, see traffic drop, then quietly redefine "success" as "more newsletter signups" because the original sales number never got written down. That is not measurement; it is storytelling.

Why do redesigns fail without a measured baseline?

Without a baseline, you cannot tell whether the new site is better, worse or the same. Traffic fluctuates for reasons that have nothing to do with your redesign — seasonality, a Google algorithm update, a competitor's price cut, a holiday. A baseline does not eliminate that noise, but it gives you a reference point. If the old site converted 2.4% of visitors into leads across the last six months, and the new site converts 2.2% across the next six, something is wrong, and you can now see it.

There is a second, less obvious failure. A redesign without a written goal lets subjective opinion drive the decisions. The stakeholder who likes blue wins the argument, not the layout that gets more people to the checkout. A conversion goal is the only thing that can settle a design disagreement with evidence instead of authority. When someone says "I think this button should be bigger," you can point at the baseline and ask: will this change make the number go up? If nobody can answer that, the change is decoration.

When do you actually need a formal goal, and when do you not?

You need a formal conversion goal when the redesign is meant to improve business outcomes — more sales, more leads, more bookings, fewer support calls. That covers almost every commercial redesign. If the site exists to make money or capture demand, the goal should be a macro conversion tied to revenue. You can still track micro conversions like page depth or video plays, but those are supporting evidence, not the headline number.

You may not need a heavyweight measurement exercise when the redesign is purely cosmetic and the site has almost no traffic — a small brochure site with 300 visits a month, for example. In that case, the cost of rigorous A/B testing or multi-variant tracking outweighs what you will learn. But you should still write down the one action that matters, even if you only check it manually every month. The discipline is cheap; the absence of it is expensive. Our team can help you decide which category your project falls into before you commit to a website redesign.

How to set a goal that survives contact with the redesign

Start with the person who owns the revenue number. Ask them one question: "If this redesign works perfectly, what single number on the site goes up?" Their answer is your macro conversion. It might be completed orders, qualified leads, demo bookings or quote requests. Write that answer down verbatim, with the date and the current value. That document is the brief.

1. Pick one primary macro conversion. It must be countable, tied to money or intent, and already happening on the current site. If it never happens today, you cannot baseline it.
2. Record the current conversion rate and volume. Use at least three months of data, ideally six, to smooth out seasonality. Note the tool you used — Google Analytics 4, your CRM, your booking system — and keep using that same tool after launch.
3. Write the goal definition in plain words. "A conversion is a completed checkout on /checkout/success with an order total above zero." Not "a sale, you know, when someone buys something." Ambiguity here makes the post-launch comparison worthless.
4. Agree on a success threshold before you see the result. Decide now: the redesign is a win if the conversion rate holds or improves by X percentage points. Decide what counts as a failure. Write both down.
5. Share the goal with everyone in the room. The designer, the developer, the content writer and the stakeholder all need the same definition. If half the team thinks the goal is "more traffic" and the other half thinks it is "more sales," the redesign will pull in two directions. The questions you ask before a web proposal should surface this tension early.

Which goal type fits your site?

Different sites optimise for different events. The table below maps the common goal types to the business model they serve. Pick one primary goal from the macro column; treat the micro column as supporting evidence, never as the headline result. A site that doubles its page views but sells nothing has failed, no matter what the dashboard says.

How do you verify the goal still works after launch?

Verification starts on launch day and continues for at least one full business cycle — typically four to eight weeks. First, confirm the tracking still fires. A redesign frequently changes URLs, button IDs or form handlers, and a conversion event that no longer fires looks like a catastrophic traffic drop when it is actually a broken tag. Check that the goal event fires on a real test conversion within the first hour after go-live, not a week later.

Second, compare like with like. The old site measured conversions on old URLs; the new site uses new ones. If your goal was "checkout complete," make sure the new checkout confirmation page is included in the same measurement view as the old one. A mismatch here — measuring the old funnel against the new one ��� produces numbers that look real and are not. Use the same tool and the same event definition you wrote down in week one, or the comparison is meaningless.

What breaks after launch, and how do you debug it?

The most common post-launch failure is a silent tracking gap. The redesign changes a form, the developer forgets to carry over the conversion event, and for three weeks the dashboard shows zero conversions while the business quietly panics. The fix is a test conversion on day one — real payment if possible, or a sandbox conversion if not — and a written checklist that includes "did the goal event fire?" as a launch gate. If the event does not fire, roll back the tracking change before you roll back the design.

A second failure mode is measurement drift. Someone redefines the goal mid-flight because the early numbers look bad. "Well, the checkout rate is down but engagement is up." That is not debugging; that is moving the goalposts. The discipline that made the baseline useful is the same discipline that makes the post-launch comparison honest. Keep the definition frozen for the full measurement window. If the number is down, investigate the funnel — where are people dropping? Which page changed? What does the old site's heatmap show that the new one does not? That is a real debugging sequence, not a dashboard refresh.

What does this cost in time and attention?

The measurable cost is mostly attention, not money. Setting a proper baseline takes a few hours of work spread over a few days: pulling historical data, writing the goal definition, agreeing on the threshold. The ongoing cost is the discipline of not changing the definition when the numbers disappoint. For most teams, the real overhead is not analytics configuration; it is the political cost of holding the line when a stakeholder wants to declare victory early or redefine failure as success.

There is a smaller operational cost if you choose to instrument events properly. A tool like Google Analytics 4 requires event setup; check the Google Analytics 4 documentation for the current event configuration options. If your site runs on a platform with built-in conversion tracking — most e-commerce platforms have this — the setup is already done and the cost is close to zero. The expensive mistake is skipping the baseline entirely, because then you pay for a redesign with no way to know if it worked, and you will likely pay for another one when this one quietly underperforms.

Common mistakes that sink a redesign

The mistakes cluster around two failures: not writing the goal down, and not sharing it. A goal that lives only in the founder's head is not a goal; it is a preference. When the designer asks why the checkout button is above the fold, the founder says "because it matters" but nobody can say how much it matters or what the current number is. The redesign proceeds on taste, and taste is a terrible project manager.

The second cluster is scope creep disguised as improvement. Halfway through the build, someone adds a new feature — a blog, a chatbot, a customer portal — that was never part of the conversion goal. The feature creep before launch is real, and it always costs the same thing: the original goal gets diluted, the launch slips, and nobody can say which change moved the number. If a feature does not serve the written goal, it waits for the next phase. That rule is what keeps a UI/UX redesign from becoming a rebuild of the whole company.

A realistic scenario: the booking site that almost redesigned itself into a hole

A tour operator in Kathmandu runs a booking site that converts 3.1% of visitors into confirmed trek reservations. The site looks dated, the founder wants a modern design, and a designer is ready to start. Before anyone opens a mockup, the team writes down the goal: confirmed bookings via the checkout page, measured in the booking system, not in analytics. The baseline is 3.1% across the last six months.

The redesign launches with a beautiful new layout, a faster checkout and a prominent "Book now" button. Four weeks in, the conversion rate is 2.4%. The number is down, and because the baseline existed, the team can see it. They dig into the funnel: the new homepage has a hero video that pushes the booking path below the fold on mobile, where two-thirds of the traffic lives. The fix is a layout change, not a redesign rollback, and the rate recovers to 3.4% by week eight. Without the baseline, the team would have celebrated the new look while revenue quietly dropped. That is the difference a written goal makes, and it is exactly the kind of work our team handles when clients come to us for a redesign with a measurable outcome.

Alternatives compared

There are three common approaches to redesign measurement, and they are not equal. The first is the one this article argues for: define a macro conversion, baseline it, redesign, compare. It costs a little time up front and pays back in evidence. The second is to redesign first and measure later — you get a number, but you have no reference point, so the number means nothing. The third is to skip measurement entirely and judge the redesign by how it looks in a stakeholder meeting. That is the cheapest option and the most expensive mistake. The simpler option is usually the right one only when the site has too little traffic to generate a meaningful baseline; in that case, write the goal down anyway and check it manually.

In short: a conversion goal before redesign is the difference between a project with evidence and a project with opinions. Write the goal, baseline it, freeze it through the build, and compare honestly after launch. If you cannot do that, do not redesign yet — measure first.

People also search for

• Questions to ask before a web proposal
• How feature creep delays a launch
• What a web development quote actually covers
• Custom software versus off-the-shelf
• Shared hosting vs VPS vs cloud

If you are planning a redesign and want the goal set properly before the first wireframe, our team can help you define the conversion, baseline the current site, and build the new one against that number. Start with a conversation about your project, or see how we approach website redesigns with measurable outcomes.