From Website Audit to Measurable Progress: How a Small Business Case Study Should Work

A useful website case study shows five things: where the site started, which problems were fixed first, who decided what, what changed over a stated period, and what else might explain the change. This article sets out that structure for a small service business using AMOS. It also shares the one documented example we can show today, our own website, and explains why it contains no customer figures yet.

Why are there no customer numbers in this case study yet?

This article contains no figures from a customer business. Results should only be published when they are real, measured over a clear period, and shared with the owner's permission.

A case study that only says "traffic went up" helps no one. You need to know what was measured, for how long, and what else happened in that time. The rest of this article shows how we would document a pilot so its results can be checked.

One rule applies to every case. AMOS does not promise rankings, traffic or revenue. It reports measured change in impressions, clicks and enquiries next to the work that was done.

Who would a typical case study be about?

AMOS is for service businesses with roughly 5 to 50 people that get new customers through their website and have no SEO, content or IT team. Examples include a dental clinic, a law firm, a bakery or a plumber.

A typical starting point might look like this:

• A dental clinic gets most new patients through its website, but no one has checked how its pages perform in Google for years.
• A law firm has strong expertise, but its practice area pages are thin and rarely updated.
• A bakery wants more custom cake orders, but its order page is hard to find on a phone.
• A plumber serves several districts, but the website does not say clearly which areas it covers.

Each of these owners knows the site could be better. The hard part is knowing what to change and finding time to do it.

How do you document the baseline?

The baseline is the starting point you compare everything against. Without it, you cannot show progress. Record it before any changes go live.

A practical baseline for a small service business includes:

1. Search impressions. An impression is one time your page appeared in search results, even if no one clicked. Google Search Console shows this for free.
2. Clicks from search. How many people actually visited from those results.
3. Enquiries. Contact form messages, calls or booking requests. Count them the same way before and after.
4. Technical findings. Problems such as slow pages, broken links or missing page titles.
5. Content gaps. Questions your customers ask that your site does not answer.

AMOS reads your pages and your competitors and shows what holds your site back, page by page, with the reason. That list becomes part of the baseline.

Write down the date range too. Four weeks of data in December means something different from four weeks in March for a bakery.

How are problems prioritised?

Most sites have more problems than an owner has time for. So the first job is choosing what to fix first. A simple way to sort them:

General guidance: start with problems on pages that already bring in enquiries. Small fixes there are easier to measure than a brand-new section.

What does AMOS do, and what does the owner decide?

The split is simple.

AMOS:

• Finds what holds your site back in Google and AI search.
• Plans and drafts fixes, articles and site changes in your brand voice.
• Shows weekly summaries of impressions, clicks and enquiries next to the work that was done.

The owner:

• Approves or rejects every fix, article and site change.
• Decides which languages to publish in. Every translation waits for approval before it goes live.
• Decides how long data is kept, and can export or delete data at any time.

AMOS does not publish anything on your website without your approval. It does not run social media or advertising, and it does not build online shops.

How do you measure before and after honestly?

Compare the same measures over periods of the same length. For example, compare four weeks before the first change with four weeks after it, and note the dates.

Then ask what else happened in that time. Common outside factors:

• Seasons. A bakery gets more enquiries before holidays whatever its website does.
• Advertising. A clinic that started paid ads in the same month cannot credit all new enquiries to website work.
• Search engine changes. Google updates can move results up or down for reasons unrelated to your site.
• Offline events. A law firm quoted in local news may see more visits that week.
• Reviews and word of mouth. These can raise enquiries without any change to search performance.

A fair case study separates the two. Changes that line up with specific work, such as more impressions for a new treatment page, can reasonably be linked to that work. Changes across the whole business, such as more phone calls overall, may have other causes. Say so plainly.

Short periods are also a limit. Search results often take weeks to reflect a change, so early numbers should be read with care.

Our own example: building softalika.com with AMOS

The one documented example we can share is our own. We built our sales website, softalika.com, with AMOS.

Between 16 and 19 October 2026 we started with the AMOS setup wizard. We then improved the site through plain-language requests to the Site Assistant. Each request became a small draft. We previewed it on desktop and phone, then approved or turned it down.

This example shows the process, not search results. Four days is far too short to measure changes in impressions or enquiries, so we do not claim any. It does show how the approval step works in practice: the owner sees each change before it goes live and can say no.

What lessons apply to any small business?

These lessons come from how the process is built, and they apply whether or not you use AMOS:

• Record the baseline first. Once changes go live, you cannot go back and measure the old site.
• Change a few things at a time. If a dental clinic rewrites ten pages in one week, it cannot tell which one helped.
• Count enquiries the same way throughout. If the plumber starts counting WhatsApp messages halfway through, the numbers stop being comparable.
• Keep a simple log. Write down what changed and when. It makes the weekly numbers easier to read.
• Reject drafts that do not sound like you. A draft that is accurate but wrong for your customers is not worth publishing.
• Decide what comes next from the data. If a new service page gets impressions but no clicks, improve its title before writing more pages.

How does an AMOS pilot work?

If you run a service business and want to document your own progress, a pilot is the place to start.

1. A short call. We talk about your website, your team and the markets you sell in.
2. Setup. The AMOS team sets AMOS up for you on your own isolated cloud resources.
3. Sign in. You sign in to the panel with an e-mail code. Only people you invite can sign in, each with a role.
4. See and approve. You see what holds your site back page by page, with the reason, and approve the fixes, articles and site changes you want.
5. Watch what changes. Weekly summaries show impressions, clicks and enquiries next to the work that was done.

The length, terms and price of the pilot are agreed on the call. There is no public price list yet.

To request a pilot or a website assessment, use the form on our contact page. Tell us your website address, your team size and what you want more of. We reply within one working day.