What Building a Local Experiences Marketplace in Morocco Taught Me About SEO and Product Design
Building a marketplace looks simple from the outside.
You have suppliers.
You have customers.
You put the offers online, add a booking button, and connect the two.
In reality, things become much more interesting when the marketplace is based on local experiences rather than physical products.
While working on MaJourneys, a platform for discovering and booking experiences across Morocco, I started noticing that travel marketplaces sit at an interesting intersection between e-commerce, local SEO, structured content, and trust.
Here are some of the lessons that stood out.
1. Your URL structure is part of the product
Imagine someone searching Google for:
quad biking in Marrakech
Another person might search:
activities in Marrakech
And another:
things to do in Agafay
These searches are related, but they represent different levels of intent.
A marketplace shouldn't force all of them onto one giant /activities page.
A structure like this is much more expressive:
/place/marrakech
/category/quad/marrakech
/category/pools/agafay
/agafay/quad/quad-biking-in-agafay
This is useful for search engines, but more importantly, it is useful for humans.
Each URL represents an actual entity or combination of entities:
Location
↓
Category
↓
Experience
That gives you a natural information architecture instead of creating hundreds of disconnected landing pages just because an SEO tool suggested some keywords.
On MaJourneys, for example, users can explore experiences across Morocco and then progressively narrow down what they want.
The SEO structure follows the browsing structure.
That distinction matters.
2. Don't build pages for keywords. Build pages for intent.
Programmatic SEO gets a bad reputation because it is easy to abuse.
A developer can generate:
/activity/marrakech
/activity/agafay
/activity/casablanca
/activity/rabat
...
and suddenly claim to have created hundreds of SEO pages.
Technically, yes.
Practically, most of those pages may provide almost no value.
The better question is:
Would this page still deserve to exist if Google didn't exist?
For a location/category page, I want the answer to be yes.
A useful destination page can explain the place, show relevant experiences, tell visitors what to expect and help them decide what to book.
For example, someone researching the Agafay area might want to compare quad biking, camel rides, pools and dinner experiences rather than immediately purchasing a single activity.
That is why we treat Agafay Desert experiences as a destination in its own right instead of merely attaching the word "Agafay" to product titles.
3. Travel has a bigger trust problem than normal e-commerce
If I buy a phone case online for $15 and it is disappointing, it is annoying.
If I book an activity during a three-day trip to Marrakech and the operator doesn't show up, I may have lost half a day of my holiday.
That changes the UX completely.
Travel marketplaces have to answer questions such as:
What exactly is included?
Where will I be picked up?
How long does the activity take?
Can beginners participate?
What happens if I cancel?
Do I pay now or later?
Those details aren't secondary content.
They are part of the product.
Take something like quad biking in Agafay.
A title and a price aren't enough.
The traveler needs to understand the duration, equipment, transfers, itinerary, safety expectations and booking conditions before clicking the reservation button.
This taught me an important marketplace lesson:
Increasing conversion isn't always about making checkout shorter. Sometimes it is about removing uncertainty before checkout begins.
4. Structured data starts with structured thinking
Developers often hear "structured data" and immediately think about JSON-LD.
But the more important structure happens before the markup.
An experience already contains relationships:
Experience
├── Location
├── Category
├── Operator
├── Price
├── Duration
├── Included items
├── Excluded items
├── Cancellation policy
├── Images
└── Booking options
Once your underlying model is clean, many other things become easier.
You can generate consistent pages.
You can create category and destination collections.
You can expose useful metadata.
You can build internal links automatically.
You can create sitemaps more intelligently.
And you can eventually provide structured information to search engines without maintaining completely separate SEO content.
In other words:
Good SEO architecture often starts as good database architecture.
5. Internal linking should model relationships
One of the most useful SEO lessons I've learned is that internal links shouldn't be inserted randomly because "SEO requires internal links."
They should describe relationships.
An Agafay quad experience naturally connects to:
Agafay
→ Quad activities in Agafay
→ Other activities in Agafay
→ Similar experiences
A Marrakech pool experience might connect to:
Marrakech
→ Pools in Marrakech
→ Day passes
→ Other Marrakech experiences
Now both users and crawlers can understand how the content fits together.
You are effectively creating a small knowledge graph around your marketplace.
6. Marketplaces have an SEO chicken-and-egg problem
A new marketplace usually starts with limited inventory.
But SEO becomes more powerful as inventory increases.
The problem is that creating thousands of empty category/location combinations is the fastest way to produce a low-quality website.
So I prefer a simpler rule:
Index pages because they are useful, not because they are technically possible to generate.
If there are enough meaningful experiences and enough useful information to satisfy the visitor, the page can become a destination.
Otherwise, it probably shouldn't exist yet.
This keeps the architecture scalable without turning the website into a collection of near-duplicate pages.
7. SEO and conversion shouldn't be separate teams mentally
Consider the content on an experience page.
An SEO person might want:
More descriptive content
Location information
FAQs
Related destinations
Internal links
A conversion specialist might want:
Clear inclusions
Transparent pricing
Cancellation details
Trust signals
Better photos
Clear calls to action
There is actually a lot of overlap.
Both are trying to answer the visitor's questions.
This has changed how I think about SEO.
I don't see it purely as optimizing pages for search engines anymore.
I see it as making the website's information architecture accurately represent what users are trying to find.
Search engines then become another consumer of that structure.
Where we're going with MaJourneys
The goal with MaJourneys is to make discovering experiences in Morocco feel less fragmented.
Instead of travelers jumping between Instagram accounts, WhatsApp conversations, hotel recommendations and dozens of individual operators, the marketplace can organize that information around destinations, categories and bookable experiences.
There is still a lot to build.
But working on it has reinforced something I think applies to many marketplaces:
The hardest part isn't displaying products. It's designing the relationships between products, places, suppliers, users and intent.
Get those relationships right and your database becomes cleaner, navigation improves, SEO becomes more natural, and users have an easier time deciding what to buy.
That is a much more interesting engineering problem than simply adding another "Book now" button.
Top comments (1)
Local marketplaces are a good reminder that SEO and product design are not separate tracks. Search brings people to a page, but the page still has to answer trust, availability, location, price, and “is this for me?” fast enough to convert.