Almost every conversation about solar panels starts from the wrong end: how many square meters of roof there are, how many modules fit, what the array costs. None of those questions determine the savings. The bill does, because a photovoltaic system doesn’t give anyone free energy: it spares you from buying it. The value of every kilowatt-hour you stop buying from CFE is printed on your bill, and not all kilowatt-hours are worth the same. This article is a guide to reading that document before you ask for a quote. The point is simple: that when you ask for one, you know what you’re asking for, and when you get it, you have the means to judge it.
The data that matters on the bill
An industrial bill carries considerably more information than most people use. In practice there are four blocks that decide whether a photovoltaic project makes sense.
Consumption in kWh. This is the energy your facility used during the billing period. It’s the number everyone looks at and often the only one. It measures quantity, not value: it tells you how much energy you used, but not what each unit cost or when you used it. On its own it isn’t enough to estimate anything.
The tariff. It’s printed on the bill and it sets the rules of the game. Some tariffs bill energy only; others bill energy and also charge for demand; and others, on top of that, split energy into time-of-use periods with different prices. Before comparing proposals, identify your tariff. Two plants with the same consumption on different tariffs do not have the same business case, and no serious proposal can be written without that piece of information.
The base, intermediate and peak periods. If your tariff is time-of-use, the bill breaks your consumption into those three blocks, each with its own price per kWh. Peak is the expensive block; base is the cheap one. The hours that define each period vary by region of the country and by season of the year, and they don’t always match intuition. That breakdown is an X-ray of your operation: without instrumenting anything, it tells you which stretches of the day your consumption lives in.
The maximum demand charge, in kW. This is a separate line from consumption and it works on a different logic. It doesn’t measure energy accumulated over the month: it measures the highest power peak the meter recorded. It’s the line most people overlook and the one that most often changes the outcome of the analysis.
With those four blocks you can already compute something far more useful than “what I pay per month”: what a kWh actually costs you in each period, and what share of your bill isn’t energy at all.
Why maximum demand changes the math
Demand is billed in kW, not kWh. You’re not being charged for what you consumed, but for the power peak the grid had to stand ready to deliver. It’s a charge for reserved capacity: the infrastructure feeding you is sized for your worst moment, and that worst moment gets paid for even if it lasted a few minutes.
Here is the trap that sinks a lot of optimistic analyses. A photovoltaic system reduces energy, but it does not guarantee a reduction in billable demand. The meter hunts for your peak. It only takes that peak occurring at night, or at the start of a shift before sunrise, or exactly when a cloud passed over the array, for the recorded demand to come out practically the same as it was before you installed the first panel. The grid has to be able to feed your entire load at any instant when the sun isn’t available, and that backup is precisely what the charge pays for.
The consequence is direct and worth stating plainly: the part of your bill that photovoltaic can actually touch is the energy component, not the demand component. When someone presents savings as a percentage of your total bill, they are folding money into the calculation that the system will never displace. There are industrial bills where the demand component weighs enough to completely change how attractive the project is. And in some time-of-use tariffs, billable demand isn’t simply the month’s peak but a figure derived from combining the demands recorded across the different periods, so that a peak occurring during peak hours weighs more than one occurring in the base period. It’s worth checking how that line is built on your bill before assuming anything.
The hourly profile: when you consume, not how much
The sun has a curve and it doesn’t negotiate: zero at night, climbing through the morning, maximum near solar noon, falling through the afternoon. Your plant has a curve of its own. The savings live in the overlap between the two.
Two plants with identical monthly kWh can get very different results. One with a daytime shift and heavy load from eleven to four self-consumes nearly everything it generates. Another with a heavy evening load or a night shift generates when it isn’t consuming and consumes when it isn’t generating. In the second case, the system exports a good share of what it produces to the grid, and that changes the arithmetic.
The underlying reason is a real asymmetry. A kWh you self-consume is worth the full tariff you would otherwise have paid, with every charge included. A kWh you export is compensated below that value, because you aren’t paid for what you pay to transmit and distribute energy. Under the distributed generation scheme, that compensation can take the form of net metering — credits in kWh, with an expiry, that are lost if you don’t use them — or net billing, settled in pesos at a market price. Neither turns surplus into a good business, and both reinforce the same design criterion: self-consumption first, export second. Which is why oversizing is throwing money away.
There’s one further detail almost nobody accounts for, and it can hurt. In much of the country the peak period falls at dusk, when the sun is already gone. In other words: photovoltaic mostly displaces base and intermediate energy — the cheap kWh — and barely touches the expensive ones. A calculation run on the average price of a kWh systematically overstates the savings, because it assumes all displaced energy is worth the same, and it isn’t.
How do you verify this in your own case? Start with the period breakdown on the bill itself: the proportion between base, intermediate and peak already tells you a great deal about where your consumption lives. If your meter has interval recording, request the demand history and work with the hourly load curve. It is the single most valuable piece of data for this decision, and it is usually available.
Twelve months, not one
A bill is a photograph. What you need is the film.
Industrial consumption is seasonal: shifts change, weather moves the air-conditioning load, production cycles have high and low seasons, plants schedule shutdowns. A single bill, taken in your best or your worst month, distorts any projection built on top of it. On top of that, the hours defining the tariff periods shift with the season, and solar generation also varies across the year, both with the height of the sun and with cloud cover during the rainy season.
Twelve months of bills let you see three things a single one hides: how your consumption moves across the year, whether your maximum demand is stable — which tells you it’s structural and not an accident — and how repeatable your profile really is. If you’re going to hand information to a supplier, hand over twelve bills. And if your plant is going to grow or change its process, say so: the bill describes the plant you have today, not the one you’ll have when the system has been running for two years.
What the bill does NOT tell you
The bill is the starting point, not the whole file. Three things determine viability and appear on no line of the invoice.
The condition of the roof. The roof deck was designed without panels on it. You have to review the dead load it will carry, and above all the wind load, which on lightweight roofs is the critical one: wind doesn’t crush the array, it sucks it upward, and that suction is far more severe at the edges and corners of the roof than at the center. You also have to review the remaining service life of the sheeting. This is the most common expensive mistake: a photovoltaic system lasts decades, so if your roof has only a few years left, somebody is going to have to dismantle the entire array to replace it. If the roof is at the end of its life, you replace the roof first.
Shading. Rooftop equipment, water tanks, stair enclosures, neighboring buildings, trees. Shade isn’t a static problem: it moves with the hour and with the season, and a shadow that doesn’t exist at noon in June can cut production on December mornings.
Interconnection capacity. The connection point, the capacity available in the transformer and the switchboards, and the physical space for equipment. The system connects to your internal network, and that network has limits. Interconnection under distributed generation also means a contract with CFE and swapping your meter for a bidirectional one, capable of recording separately what you consume and what you deliver. Not every site can accommodate, on the electrical side, the array it can accommodate on the roof.
If by the time you finish reading your bill you know which tariff you’re on, what share of your invoice is demand, and which periods your consumption lives in, you’re equipped to hold a genuinely technical conversation and to spot a proposal built on a napkin calculation. That’s the goal. At De La Mar we design and install industrial solar panels starting from twelve months of bills and a visit to the roof, never from a catalog. If you’d like us to review yours before you decide anything, get in touch and we’ll go through the numbers.