TLDR: The brokerage hired me to grow its website, and the first document I handed the owner was a list of twenty-six pages to remove or merge. The audit had to come before any strategy, because you cannot plan what to build on top of pages that cancel each other out. Getting sign-off from the people attached to those pages took longer than finding them, and the smaller site that resulted was the operational win: fewer pages to keep accurate, fewer to translate, one page per job.
The owner had expected a plan. He had cleared an hour, he had a notebook open, and he was ready to hear about the pages we would build. What I put on the table instead was a single sheet with two columns: the page, and what should happen to it. Twenty-six of the brokerage's sixty pages were on it. Next to each one was either "merge into" and a destination, or "remove".
He read it twice. Then he asked the question every consultant should be ready for on day one: "We hired you to make the site bigger. Why is your first idea to make it smaller?"
I work in-house on the digital side of a luxury real estate brokerage in Marrakech, alongside a head of sales and a small team of agents. The answer I gave him that morning is the same one I would give any business whose website has grown by accretion for years: the site was too crowded, and the crowd was hiding the pages that mattered.
Why the Audit Came Before the Strategy
There is a version of consulting where the first deliverable is a vision. It is easier to sell and pleasanter to present. I did not have the option, because I did not yet know what the website was. Nobody did. Pages had been added by three different agencies and two former employees over several years, and no one person had a list of them.
So the first six weeks were an inventory. Every page, its purpose, what it was meant to rank for, whether it had any inventory attached, whether Google was showing it to anyone at all, and whether another page on the site was trying to do the same job. Ahrefs describes this kind of content audit as deciding, page by page, whether to keep, improve, merge or remove, and that is exactly the decision I ended up making sixty times.
The strategy could not come first because the strategy depends on the answer. If the site already had a good page on foreign ownership, the plan was to improve it. If it had three weak ones, the plan was to merge them. If it had none, the plan was to build one. Those are three different plans, and only the inventory tells you which one you are in.
How the Twenty-Six Were Found
The twenty-six invisible pages fell into three groups, and each group needed a different piece of evidence before anyone would believe it.
The first group was duplicated neighbourhood pages. The same area existed under two or three different names and spellings, each page written at a different time by a different person, each saying roughly the same thing. Google had picked one to show and ignored the others, which is what it does when it finds duplicate URLs for the same content. The evidence was simple: the search console showed one page with impressions and its siblings with none.
The second group was property-type pages that repeated each other. Villas, luxury villas, villas with pool, villas for sale: four pages, one intent, and the same dozen properties on all of them. The evidence here was the listings themselves. When four pages share one inventory, they are one page with four addresses.
The third group was the thin area guides. Pages of two paragraphs about a district, written to exist rather than to inform, with nothing a buyer abroad could act on: no distance to the airport, no note on the international schools, no comparison to the neighbouring district, no properties. The evidence was the buyer questions I had collected from two years of first emails, none of which those pages could answer.
Together the three groups came to twenty-six pages that Google was effectively not showing to anyone. Removing them would lose nothing, because they were already invisible. Merging them would give the surviving page the text, the inventory and the internal links of its siblings.
Getting Sign-Off From the People Attached to the Pages
Finding the pages took six weeks. Getting permission to touch them took longer, and this is the part of the deliverable I had underestimated.
Every one of those pages had a parent. One of the agents had written the original page for the district he grew up in and did not want it folded into a page about the wider area. A former marketing hire had built the property-type pages, and the owner remembered paying for them. The thin area guides had been the brokerage's first attempt at content, years earlier, and there was a sentimental attachment to having "covered" every district, even in two paragraphs.
I made three mistakes in the first conversations and learned from each.
- I led with the search data. Impressions and indexing status meant nothing to an agent who had written the page and knew it was accurate. Accurate was never the question.
- I presented the list as a whole. Twenty-six removals in one sheet reads as a verdict on everyone's past work at once. Nobody signs a verdict.
- I used the word "delete". For the owner, delete sounded like waste. Merge sounded like an upgrade, and in most cases merge was the truth.
The conversations that worked went page by page, with the person who cared about that page, and they started with the buyer rather than with the data. For the agent's district page, I asked him what a buyer from Geneva would need to know about that district before choosing it over the next one, and we agreed that the answer was longer than his page and belonged on the merged one, with his name on it as the source. For the owner's property-type pages, I showed him the same twelve villas on four pages and asked which one he would send to a buyer. He picked one. That was the sign-off.
The thin area guides were the easiest in the end. Once the owner had seen what a proper area guide looked like, with distances, schools, the honest comparison and the inventory, he asked why we had ever published the short ones. Redirecting each old address to its new home, using a plain 301 redirect, meant no bookmark and no old link from anywhere else broke, which was the last objection anyone had.
Why the Smaller Site Was the Efficiency Win
The owner's worry was that fewer pages meant less presence. What he saw within a quarter was that fewer pages meant less work, and better work.
Every page on a real estate brokerage's website carries a maintenance cost that nobody budgets for. Its inventory has to be current, or a buyer sees a villa that sold in the spring. Its distances and school notes have to be checked when a road opens or a school moves. And because our buyers search in French as well as English, and the site had been English-only, every page we kept was a page we would have to write properly in French. Sixty pages to maintain and translate is a job for a team we did not have. Thirty-four was a job I could plan.
The consolidation also gave every surviving page one job. The merged neighbourhood page holds all the inventory for that area, all the internal links that used to be split across its siblings, and the agent's knowledge of the district. A buyer who lands on it does not have to guess which of three pages is the real one, and the agents do not have to guess which page to send. The head of sales noticed that last part before anyone else did: sending a buyer a link had become a decision that took no thought.
The removal list was, in the end, what made the build list credible. When I came back with the keyword map, which pointed to several hundred pages worth creating over time and 126 topic clusters the brokerage had never covered, the owner did not flinch, because he had already seen that I was willing to take pages away. A consultant who only ever adds is easy to doubt. One who has just merged twenty-six pages on the brokerage's website has earned the right to propose the next hundred.
What I Would Do Differently
I would still start with the audit. I would still hand over the removal list before the build list. What I would change is the order of the conversations. I would talk to the people attached to the pages before I finished the sheet, so that the deliverable arrived with their names already on the merges and the sign-off already half done. The list would have been shorter to present and quicker to approve, and I would have spent the weeks I lost on the difficult ones building the first of the pages we needed.
The owner still has the sheet. He keeps it in the same notebook he had open that morning, the one he had brought to write down a plan.
