Valentina's grandmother had arrived in Glasgow in 1954 with a cardboard suitcase, a recipe for pistachio gelato that she refused to write down because written things could be stolen, and a determination that had never been fully captured in English because English did not have the right word for it.
She had come with the wave of southern Italians who understood that ice cream was a thing the Scottish wanted and were not making properly, and who had built an entire community in Glasgow on that understanding. Valentina's grandmother had not opened a shop. She had worked in other people's shops until she understood the city well enough to understand what it actually needed, which was not ice cream in January but something for the table at weddings and communions and the particular Scottish occasions that required a great deal of food prepared by someone who took it seriously.
The catering business had grown across three generations in the way family businesses grow. not by strategy but by reputation, each generation inheriting clients from the last and adding their own. By the time Valentina took it over it was feeding two hundred people on a good Saturday and had a waiting list for summer weddings that stretched eighteen months out.
The website had been built remotely by her cousin in Reggio Calabria, who was a graphic designer and who had done it over a series of video calls where the connection kept dropping and the brief kept changing because Valentina's mother kept joining the calls to make suggestions. It was finished eventually. It looked like a catering website. It had photographs of the food, which were excellent, and a contact form, and a page about the family history that Valentina's mother had written and that nobody had edited because editing it would have caused a problem nobody wanted.
It loaded in eight seconds on mobile.
The Google Ads campaign had been running for fourteen months. It targeted wedding catering keywords in Glasgow and the surrounding area. It had a budget of three hundred pounds a month. It had produced eleven enquiries, of which four had become bookings, of which two had been for events small enough that the margin barely covered the cost of the campaign.
Valentina's grandmother, who was ninety-one and still made the gelato herself on Thursday mornings and who had never in her life owned a mobile phone or heard the phrase Quality Score, had in the same fourteen months personally referred six bookings worth significantly more than anything the campaign had produced.
Valentina knew this. She had done the maths.
She had not yet decided what to do about it.
The Auditor's Take
Names and identifying details have been changed. The technical profile below is based on real observations from real sites. The diagnosis is always free.
An eight-second mobile load time on a wedding catering website carries a specific cost in a specific market. Couples planning a wedding search on phones, often together, often in the evening, often while looking at multiple options simultaneously. The decision window for a catering enquiry is narrow. A page that takes eight seconds to show them the food photographs. the one thing that will either move them to enquire or not. loses them before the photographs appear.
The campaign's eleven enquiries from fourteen months of spend is not a targeting failure. The keywords were appropriate. The geography was correct. The issue is the conversion rate from click to enquiry, which was suppressed by the load time and further suppressed by a contact form that was, on closer inspection, sending submissions to an email address that Valentina's cousin in Reggio Calabria had used during the build and never changed to Valentina's actual address.
Some of the people who clicked the ad and filled out the form never heard back. Not because Valentina ignored them. Because the form was sending their enquiries to Calabria.
The grandmother's referrals arrived by phone. Directly. Without a form. Without a page load. Without a cousin in another country in the delivery chain.
The Technical Tax in this case had two components. A slow page that cost the campaign its conversion rate. And a broken form that cost the campaign its leads. Together they turned three hundred pounds a month into a fraction of what it should have produced, while a ninety-one-year-old woman with a pistachio gelato recipe and no smartphone outperformed it by direct comparison.
The lesson is not that digital advertising does not work. The lesson is that it requires the same attention to detail that the grandmother applied to the recipe she refused to write down.
Based on true events. All names changed. The Auditor's Take is a regular column on vsourcecode.com documenting real patterns from real sites. The diagnosis is always free.
Ready to Fix This on Your Site?
Run your site through PageSpeed Insights and send me the score. Free diagnosis. No pitch. Just the numbers.