How to Audit Your Current Website Before Hiring an Agency

Most people walk into the first agency call knowing only that their website feels wrong. It looks dated, or it feels slow, or enquiries have gone quiet, and that vague unease is the whole brief. The trouble with a vague brief is that it hands the agency the job of telling you what is wrong with your own site, and an agency describing problems it has not been challenged on can describe whatever happens to match the services it most wants to sell. Half an hour of your own time, before you talk to anyone, changes that completely.

So here is how to audit your website before hiring an agency. Spend roughly thirty minutes running six plain checks. Run PageSpeed Insights and write down the speed numbers. Open the site on your own phone and look at it honestly. Count your pages and find the dead ones. Submit your own contact form and confirm the lead actually arrives somewhere. Confirm analytics is genuinely installed. Confirm that you, and not a former developer, control the domain and the hosting. Then write the findings on a single page and carry that page into every agency conversation. You do not need to be technical to do any of this, and thirty minutes really is enough, because the goal is not to fix the site. The goal is to walk in informed.

Check the speed: PageSpeed Insights and the two numbers that matter

Start with speed, because it is the easiest thing to measure objectively and the easiest thing for an agency to be vague about. Go to pagespeed.web.dev, paste in your homepage address, and run the test for both mobile and desktop. Then run it again on one or two of your important inner pages, a key service page or your main product page, because the homepage is often the most cared-for page on the site and not representative of the rest.

Ignore the big score out of one hundred. It moves around and it invites arguing. Instead, write down two real numbers. The first is Largest Contentful Paint, usually shown as LCP, which measures how long it takes for the main content of the page to appear. Under two and a half seconds is good. The second is Interaction to Next Paint, shown as INP, which measures how quickly the page responds when someone taps or clicks. Under two hundred milliseconds is good. Note those figures for mobile and desktop. They are objective, they are Google's own measures, and they mean an agency cannot tell you the site is fine when it is not, or tell you it is broken when it is merely a little slow.

Check the phone: how the site really behaves where your customers are

The next check costs nothing and most owners skip it because they always look at their own site on a large desktop screen. Your customers do not. Most of the people visiting your website are on a phone, so pick up your actual phone and look at the site the way they do.

Scroll through every page that matters. Look for text too small to read comfortably, buttons crowded so close together that the wrong one is easy to hit, images that spill off the edge of the screen, a menu that will not open, a strip of horizontal scrolling that should not be there, a pop-up that cannot be closed with a thumb. Then try to actually do something. Find a price. Book a call. Start to buy. Pay attention to the moments where it felt awkward or where you had to pinch and zoom, and note them down. That list is not a complaint. It is evidence, and it tells an agency exactly where the experience is failing the people you most want to reach.

Check the pages: count them, find the dead ones

Now get a clear picture of how big the site actually is. Open your main navigation and write down every page. Click into the menus, follow the footer links, and note anything that exists but is not linked from anywhere obvious. Owners are often surprised here, because a site that felt small turns out to have thirty pages, or a site that felt substantial turns out to be six.

As you go, mark the dead pages. A dead page is one that no longer earns its place: a team page listing people who left two years ago, a services page for something you no longer offer, a blog with its most recent post dated 2022, an events page advertising an event long past. Note, too, the pages that should exist and do not. You are not fixing anything in this check. You are building an honest inventory, so that when an agency quotes for a redesign it is quoting against the real site rather than a rough guess, and so you can decide deliberately what to keep, what to cut and what to write fresh.

Check the form: the test almost nobody runs

This is the check that almost nobody performs and it is the one that quietly costs the most money. Go to your contact form, fill it in exactly as a customer would, and submit it. Then do the part that matters: confirm the enquiry actually arrived somewhere a human being will see it. Check the inbox. Check the CRM. Wherever leads are supposed to land, confirm yours landed there.

While you are at it, confirm that inbox or account is one somebody still watches. Plenty of forms quietly send to an address that belonged to a staff member who left, or to a folder nobody has opened in a year. A contact form that silently fails is not a small cosmetic flaw. It is a website losing you real enquiries every single day, and it is the kind of problem that is far cheaper to find now, before a redesign, than to discover months later when you wonder where the work went. If the form does fail, that single finding can justify the whole exercise.

Check the foundations: analytics, domain and hosting ownership

The last two checks are about the foundations underneath the site, and they are the ones an honest agency most wants you to have done.

First, analytics. You cannot brief an agency properly on a problem you have never measured, so confirm that Google Analytics 4, or any analytics tool, is genuinely running. A quick way is to install the Google Tag Assistant browser extension and load your site, or to view the page source and search it for a tag such as gtag. If nothing is there, that is a finding worth writing down, because it means nobody has data on how the site performs. If analytics is installed, log in and note roughly how much traffic the site gets and which pages get most of it. That picture should shape any redesign.

Second, and most important of all, confirm that you control the domain and the hosting. This has nothing to do with how the site looks and everything to do with whether a new site can be launched at all. Can you log in to the domain registrar where your web address is registered? Is the domain held in your company's name, or in a former developer's personal account? Do you have the login for the hosting account? If a past developer or agency holds these credentials and you cannot reach them, that is a serious problem, and an agency needs to know it on day one rather than discover it the week of launch.

The one page you take to the agency call

The output of the thirty minutes is a single sheet of paper, and it is the most useful thing you can bring to an agency meeting. Write down the PageSpeed numbers, the LCP and INP for mobile and desktop, for the homepage and a key inner page. Add a short note on the mobile problems you found. List the page count and the dead pages. Record the result of the contact form test and where leads currently go. State whether analytics is installed and roughly how much traffic the site gets. Note the domain and hosting ownership position, flagging any access you do not currently have. Finish with two or three plain sentences on what you actually want the new site to do, whether that is generating more enquiries, supporting a new service, or simply not embarrassing you in front of customers.

That page does three quiet but powerful things. It turns a vague feeling into a concrete brief. It lets every agency quote against the same set of facts, which is the only way the quotes you receive are genuinely comparable. And it signals, without you having to say so, that you are an informed buyer, which tends to change the entire tone of how an agency talks to you and what it tries to sell.

When the self-audit finds something real

There are two honest ways this can go. If the self-audit comes back clean, the speed numbers are fine, the phone experience is smooth, the form works, the foundations are in order, then you may not need a full redesign at all, and you should be wary of any agency that insists otherwise. A clean audit is a perfectly good outcome.

More often, though, the thirty minutes turns up something real: a homepage taking five seconds to show its content, a contact form that has been failing silently, a domain you do not control. When that happens, it is worth knowing what a self-audit cannot do. Half an hour by a non-developer finds the obvious, visible problems, and that is genuinely valuable, but it cannot see the structural ones underneath. It cannot tell you why the page is slow, whether the code can be safely extended or needs replacing, where the security exposure sits, or how much hard-won search ranking a careless rebuild would throw away. Those questions need a deeper, professional audit.

That is the work we do at WitsCode. When your own thirty-minute check has surfaced real problems, a proper technical audit explains the cause behind each one, separates what genuinely needs rebuilding from what can simply be fixed, and gives you a clear, costed picture before you commit to anything. Run the self-audit first, because it is yours to keep whatever you decide next. If it finds something that matters, bring that one page to us and we will tell you what is actually going on underneath.