Billboards: the sites that run at a loss

“I have forty sites, and for six years I thought the worst of them simply brought in a little. It turns out it brings in minus 27 500 ₴ a year. And I have twelve of those.”

Zoriana Boichuk runs an outdoor advertising network in Ivano-Frankivsk. Forty 3 × 6 boards, from the central junctions out to the exit roads on the edge of town. Nine years in the business, six of them as the owner. The network is profitable, the clients are regulars, there are no debts. And inside it twelve sites work all year round taking money out.

A network of forty sites

Outdoor advertising looks like a simple business. There is a board, there is a client, there is a price per month. The difficulty is not in the selling. It is that every site is a separate little business with its own economics, while the cash flow for all of them is one.

Revenue arrives by client. Costs sit by site. The report piles both into the same heap.

• ground rent for the site: 38 000 ₴ a year
• depreciation of the structure itself: 9 000 ₴ a year
• fixed cost per board, total: 47 000 ₴ a year
• printing and hanging a poster: 2 400 ₴ a change

The first three lines do not care whether the board is sold. The fourth appears only when there is a client.

That is exactly why outdoor advertising is so easy to misread from the inside. The costs here are almost entirely fixed and the income almost entirely variable — and any decision about a location runs for a year, because a ground lease never comes shorter.

What outdoor advertising actually sells

The product here is not the board. It is a month on the board.

An unsold month never comes back

A hotel can make up its occupancy next week. A shop can clear the leftovers at a discount. Outdoor advertising can do nothing at all: if a board stood empty, that month is simply gone, and you cannot sell it later for a single hryvnia.

Which is why an empty site does not cost zero. It costs its share of the ground rent, already paid up front.

The client pays to be seen. The network sells time, and time does not go into stock.

The site-month as a unit

Forty sites across twelve months give 480 site-months a year. That is the whole capacity of the network, and physically there is nothing beyond it to sell.

Last year 313 were sold.

The unit matters here, and not for the sake of reporting. As long as capacity is measured in sites, forty boards look like identical assets. The moment it is measured in site-months, you can see that one board sells eleven units a year and another five and a half.

Occupancy on its own says nothing

313 out of 480 is a bit over 65%. For outdoor advertising in a regional capital the figure sounds decent. But it is made of three very different parts, and the average between them has no practical meaning whatsoever.

Three groups of locations

The centre is occupied almost all the time. The avenues stand empty a third of the year. The outskirts are unsold for most of the year, and Zoriana had known that for all six years.

That is the trap of the average. The 65.2% of occupancy describes no real site in the network: not a central one at 91.7%, not an outlying one at 45.1%. The average exists only in the report.

“Right up to the end I thought the outskirts were just a weak segment. Lower price, lower occupancy, smaller income. It never crossed my mind that the income there could be negative.”

What the year brings in

10.3% net. For a network with no loans and with every structure owned outright that is a respectable result, and Zoriana was happy with it two years running.

The second line is worth a look. Half the revenue goes on ground rent and depreciation — 50.3%, the largest item in the whole business. Printing takes 20.1%, office and sales another 19.3%. Profit gets whatever is left.

The report shows how much the network earned. It does not show who inside it earns.

Where the loss hides

The problem is the 1 880 000 ₴ line. That is forty payments of 47 000 ₴ each, sitting in the books as a single sum. While they sit together, no individual site has either income or costs — it is just part of a network.

That is how any bookkeeping works: costs are grouped by their nature, not by the object that creates them. Rent sits with rent, printing with printing, depreciation on its own. For the tax office that is correct. For decisions it is useless.

Zoriana checked whether this was a quirk of her own accounts. It was not: none of the three programs she tried over six years offers a site as a dimension for costs. There are categories, there are counterparties, there are projects. The object is not among them — you have to set it up by hand.

Rent does not shrink when a board stands empty

The landowner takes no interest in whether the board is sold. The lease is signed for a year, the payment goes up front, and the 38 000 ₴ leaves all the same — for a site that is occupied eleven months and for one that is empty for almost seven.

What one site contributes

The calculation is simple. Take the group's revenue, subtract rent with depreciation and printing, divide by the number of sites.

The right-hand column holds the most important number in the table. It is the occupancy at which a site stops losing money: a central board needs 25.1%, a board on an avenue 45.5%.

Contribution looks at the location, not at the client

One central site brings 124 600 ₴ a year. One on an avenue brings 21 800 ₴, which is 5.72 times less. One on the outskirts takes 27 500 ₴ away.

The gap between the centre and an avenue is almost sixfold, although the price differs by only 1.64 times. All of the gap is made by occupancy, because the ground rent is the same for all forty sites.

And the outlying board needs 108.8%, which is impossible by definition.

The same thing can be said in months. A central board pays back its 47 000 ₴ after 3.01 months of selling, a board on an avenue after 5.47. An outlying one needs 13.06 months, and a year holds twelve.

The outskirts

Twelve sites at minus 27 500 ₴ each add up to minus 330 000 ₴ a year.

The reason is not bad salespeople. A month on an outlying board costs 6 000 ₴, printing takes 2 400 ₴, so the contribution is 3 600 ₴. Twelve such months give 43 200 ₴ — which is 3 800 ₴ short of the site's fixed costs.

A board at 6 000 ₴ does not pay for itself even when it is sold the whole year.

To cover 47 000 ₴ over twelve months, a site has to deliver 3 916.67 ₴ of contribution every month. With printing on top that is a price of 6 317 ₴. The outskirts sell at 6 000 ₴, which is 317 ₴ below the line where selling makes any sense at all.

The network's net profit for the year is 382 800 ₴. The outskirts eat 86% of it.

Without the twelve worst boards the network would earn almost twice as much.

In figures that would be 712 800 ₴ instead of 382 800 ₴. It was precisely that difference that had Zoriana phoning the installers the same evening.

“When I saw the table, my first reaction was — we take them down tomorrow. Twelve poles, two days of work, and I am three hundred and thirty thousand up. I was already calling the installers.”

Why they were never taken down

For six years nobody counted the contribution of an individual site. Not out of laziness — that cut simply does not appear by itself.

• in the books revenue sits by client, not by board
• ground rent arrives as a single payment for all forty sites
• printing is paid for in batches, not poster by poster
• depreciation does not even look like a cost, because it demands no cash

There is one more reason, a purely human one. The outskirts were once the most promising direction: cheap rent, new districts going up, a bypass road promised out that way. The bypass never came. The twelve sites stayed.

To see that minus 27 500 ₴, all four lines have to be broken out by site. Not one of those operations is difficult — it is just that nobody did them.

Ground rent is paid up front

A site costs the same whether it sells or not

Ground rent goes out as one payment, a year in advance. That is 1 520 000 ₴ for forty sites, handed over before the network has sold a single site-month. At that moment all the boards are identical, and the worst of them differs from the best in nothing.

The sum leaves the account in one go, and in that month it looks like a catastrophe. Then nobody mentions it for eleven months, because in the reports it is smeared out evenly.

The difference shows up later, over the year, and it shows up only in occupancy.

An empty month cannot be sold

The outskirts stand empty for 79 site-months out of 144. Every one of them has already been paid for in rent and will never bring in anything.

This is not a lost opportunity. It is paid-for capacity left unused, and the difference between those two ideas is the whole substance of outdoor advertising.

“I once tried selling an outlying board at 3 000 ₴, just to get something. The client took it for two months. And then I realised I had covered the printing and a little over, and still paid the full rent for the site.”

The arithmetic of taking them down

Zoriana started counting and ran into something the table did not hold.

Three clients bought packages. They wanted the centre, and for the centre they paid full price, but they agreed only on condition that cheap boards on the exit roads came with it — coverage for a delivery operation and for two car dealerships.

“They told me straight out: without the outskirts we are not interested, we will go to the competitor who has the whole route. For three years I was selling the centre through the outskirts and I could not see it.”

The packages that hold the loss in place

A package looks like a discount and works like a subsidy

Inside a package the outskirts go at list price, not cheap. The loss there comes not from the price but from the fact that the boards stand empty the rest of the year. A package client covers two or three months out of twelve, and that is not enough to earn back 47 000 ₴.

But while he does it, he is paying for the centre. And if the outskirts disappear, he leaves the centre too.

The rule here is simple, if unpleasant. If a cheap position in a package does not cover its own 47 000 ₴ for the year, it is not the client paying for it — it is the central boards. That does not make the package bad. It just means you count it on contribution, not on the size of the invoice.

Between them the three clients hold 20 site-months in the centre.

The line where the move turns into a loss

The contribution of one central site-month is 18 000 ₴ of price minus 2 400 ₴ of printing, so 15 600 ₴. The ground rent in the centre is annual and does not vanish because a client walked away.

Twenty site-months at 15 600 ₴ is 312 000 ₴ of contribution lost.

• remove 12 outlying sites: +330 000 ₴
• lose 20 central site-months: −312 000 ₴
• net for the year: +18 000 ₴
• that is 4.7% of the annual net profit

The break-even line is counted in site-months

Divide 330 000 ₴ of saving by 15 600 ₴ of contribution and you get 22 site-months. So if the package clients take away 22 site-months instead of twenty, the move turns into a loss. The headroom is two site-months, or one bad phone call.

Two site-months of headroom is not a strategy, it is a coincidence. Any fourth package client, any change of route at a logistics firm, any dealership deciding to cover one more exit road, and the plus eighteen thousand becomes a minus.

What Zoriana does now

“I am not taking the outskirts down. The arithmetic says it gives me eighteen thousand a year at the risk of losing a client who pays for the centre. But now I know the price of every board, and with that number the conversation with a client sounds different.”

The most painful part turned out to be not the figure but the term. A lease on an outlying site is signed for a year, and you cannot get out of it early without a penalty. So a decision Zoriana wanted to take in two days physically stretches over twelve months.

Instead of taking boards down the network does three things. The outskirts no longer go into packages for free — they are sold as a separate position with a price of their own. The two worst sites were not re-signed for the next year, because their leases were running out anyway. And every new package is now counted on contribution, not on the size of the invoice.

• the contribution of every site is recalculated once a quarter
• no package is approved if its total contribution is below 47 000 ₴ per board
• leases on the outskirts are not renewed automatically
• new locations are taken only where the price a month is above 6 317 ₴

The fourth decision is the quietest of them all. Zoriana stopped measuring the network's work by occupancy and measures it by contribution — and that changed which locations look attractive to her in the first place. A cheap site with cheap rent no longer looks like a bargain.

None of these decisions needed money. All of them became possible only after the 1 880 000 ₴ of rent was broken out into forty lines.

The money does not vanish — you simply do not see it

“The strangest feeling was not about the minus twenty-seven thousand. It was that I have been nine years in this business and nobody ever told me you can count contribution board by board. Everybody counts occupancy.”

The forty sites still stand where they stood. Twelve of them are still in the red. The difference is that the red now has a sum, a term and a cause, instead of a reputation as a weak segment.

Zoriana's network did not get bigger and did not raise its prices. The revenue stayed 3 734 000 ₴, there are still forty sites, the occupancy is the same. What changed is that every decision about a location or a package is now taken with the contribution figure in front of her, rather than with a feeling that the outskirts are just a weak segment.

Arithmetic does not take decisions. It only shows you what it is you are deciding.

How to see this across your own network

All the arithmetic above rests on one cut the cash report does not make: revenue and fixed costs not in total, but per site. While the ground rent sits as a single payment for forty locations, twelve loss-making sites hide behind eight in the centre — and the network keeps for years the thing that is eating it. When income and costs break down by site on their own, a loss-making location shows up in the second month — which is exactly why owners of small and medium businesses use Finmap: to see the financial picture instead of piecing it together.

Read also

• Tire Service: Seasonal Profit Per Bay, Not Just a Busy Season
• Scenario Planning for an Agency: Three Versions of Next Quarter

Topic foundation: Management Accounting Explained: What It Is and Why Owners Need It