Case 03 · Events venue, B2C and B2B · three years
A venue renting space for private and corporate events. Three years on the account, including the point where everything that had been bringing customers in simply stopped.
Ratios and percentages only. Absolute spend and revenue belong to the client rather than to my marketing, so those stay for the call.
The channels that had been producing customers stopped producing them, and not gradually. This was not a bidding problem or a creative problem. The demand sources themselves had gone, and no amount of optimising inside the existing accounts was going to bring them back.
The tempting move is to start testing new channels immediately and hope one lands. That burns budget in a straight line and usually teaches nothing.
Instead of guessing at replacements I went into the CRM history to find where customers had actually come from over the preceding years, including the sources nobody was actively buying. That is a research job rather than an advertising job, and it is the reason the answer was findable at all.
The history surfaced a category of marketplace that businesses like this one sit on. They had listings there and had never bought advertising on any of them in any systematic way. I tested all of them over roughly two months, found the one that produced genuine customers, then bought every placement it had, because almost no competitor was bidding there and the prices reflected that.
Customers were segmented by occasion: birthdays, corporate events, team building, training. The useful discovery was lead time. Bookings are made one to two months before the event on average, but it varies sharply by occasion, and the most profitable corporate bookings for New Year are decided two to three months ahead. That means the offer has to be live and visible by the end of August, not in December when the event happens.
Each occasion got its own window and its own offer, and the calendar was built from the decision date rather than the event date.
Low competition and low prices, which is what an unbought channel looks like before everyone finds it. It produced volume the business had never seen from paid, and it is the single reason that year recovered.
Where the occasion-specific offers actually landed. The conversion gain came from rebuilding what was being offered and to whom, timed to the decision window, rather than from redesigning the page. Layout was not the lever here.
Bookings of this size get discussed on the phone, so calls were tracked back to source. Without that, the channel actually producing revenue would have looked like the channel producing nothing.
Return on ad spend of 10.9 across the engagement, company revenue up 250%, and paid acquisition growing into around 30% of the whole business.
The transferable part is not the channel, because that particular marketplace belongs to that category. It is the method: read the customer history before buying anything, look for demand nobody is paying for yet, and build the calendar around when people decide rather than when they show up.
The first step is the same either way: a paid diagnosis that works out what you can actually afford to pay for a customer, and tells you honestly whether ads are your bottleneck.