SEO for Solar Companies · Guide

How to Present ROI and Savings Estimates Responsibly

A savings tool converts better than anything else on a solar site, because it answers the question everybody arrives with. It is risky for precisely the same reason. The job is keeping the thing that works without making a claim the firm cannot support.

Updated: August 2026
Written by: Andrew Odgers, Managing Director
Reading time: 11 minutes
Concede the commercial point first

It Answers The Only Question That Matters

Customers want to know what this means for them. A tool offering that answer converts better than anything else a solar firm can put on a website. Any argument that starts by denying it will be ignored.

Why it works so well. It is the question.

Everything else on a solar site is preamble to whether it is worth it. A tool promising to settle that in ninety seconds is offering the one thing the visitor came for.

Why the appeal is stronger here than elsewhere. The reader cannot work it out.

Somebody buying most things can estimate the value themselves. Here the arithmetic depends on their consumption, their tariff and their roof. They know they cannot do it.

What this page is not arguing. That you should remove it.

The tool is valuable and the instinct to have one is correct. What follows is about the difference between a version that helps and a version that creates a liability.

Our own position, stated up front. We do not build one.

No page in this cluster contains a calculator. Our format bars interactive elements below the hero. Separately, a savings tool would itself be a claim. Both reasons hold on their own.

The block that earns the page

And That Answer Is A Claim

The moment a firm's tool produces a figure, that figure becomes something the customer relied on. It is not a service, an illustration or a bit of content. It is a statement the firm made.

Why the framing matters. Intention is irrelevant.

A firm producing a number thinks of it as a rough guide. The customer receiving it thinks of it as what the company told them. Where those diverge later, the customer's version is the one that gets repeated.

What the tool cannot know. Almost everything that matters.

How much electricity the household uses, when they use it, what they pay, what they will pay next year and how the roof actually performs. A calculator asks for a fraction of that and assumes the rest.

Where this sector got into difficulty. Precisely here.

The complaints this industry is remembered for turn on customers who were given figures and did not get them. That is the mechanism. A calculator is the modern version of it.

What follows. Not removal. Transparency.

A number with its workings shown is defensible. A number on its own is not, which is what blocks three and four are about.

Wording, plus where it sits

What The Output Must Say

Three things, in plain language, positioned where somebody will actually read them.

That it is illustrative. The first word, not the last.

An estimate rather than a quotation or a projection. The distinction is real and customers understand it perfectly well when it is stated rather than buried.

That it is based on stated assumptions. Which are shown.

Not that assumptions were made, which sounds evasive. That here they are. Block four is entirely about this because it is the recommendation that matters most.

That it is not a guarantee. Said once, plainly.

Not in a paragraph of qualifications. One sentence, in the same size text as the number.

Where it goes. Beside the figure.

Alongside the answer rather than in a footnote, a tooltip or terms nobody opens. Wording positioned where it will not be read is not a qualification. A firm relying on it has not qualified anything.

Why this does not deter. It reassures.

A customer who understands what an estimate is remains interested. One who later feels a number was oversold does not.

The single most important recommendation

Show The Assumptions, Not Just The Answer

A tool that displays the electricity price, the export rate, the usage pattern and the generation estimate it used is transparent, defensible and considerably more persuasive than one that shows a figure alone.

Why it is more persuasive rather than less. The reader can check themselves.

Somebody who sees the assumed electricity price can compare it with their own bill. If it matches, the number gains credibility it could never have had on its own. If it does not, they adjust and get something useful.

Why hiding them backfires. Sophistication.

This buyer is wary and frequently well informed. A bare figure invites the question of what produced it. A tool that will not say reads as one that would rather not be examined.

What showing them does legally. Changes what was said.

An estimate with visible inputs is a conditional statement rather than an assertion. That is a materially different thing to have published. It is the difference between an illustration and a claim.

The practical form. Editable, ideally.

Assumptions a reader can change are better still, because the output becomes theirs rather than the firm's. That is the version described in block ten.

Why a single headline figure misleads

The Assumptions That Move The Answer Most

Four inputs do most of the work. Small changes to any of them change the result enormously. That is why one number presented as the answer is misleading even when it was calculated carefully.

Consumption pattern. The largest single factor.

Whether the household is using electricity while generation is happening. The installation page covers why this matters more than sunshine. It is also the input most tools ask about least.

Tariff. What displaced consumption is worth.

The rate a household pays decides the value of everything the system offsets. It is also something the customer can change next month.

Export rate. What the unused portion is worth.

Rates differ between suppliers and the customer chooses. A tool assuming one has assumed a decision the household has not made.

Shading. The physical variable.

What actually falls on the roof across a day and a year, which no tool can know from a postcode. The others are described in SEO for solar panel installation.

Stated plainly

Never Present A Best Case As Typical

Inputs should be conservative. An optimistic default produces a misleading output whether or not anybody intended it, because the reader has no way of knowing the dial was set generously.

How it happens without malice. Every choice leans one way.

A slightly high electricity price, a slightly favourable usage pattern, a slightly optimistic generation figure. None is misleading alone. Together they produce a number nobody will achieve.

The asymmetry that settles it. The two errors are not equal.

A firm whose estimate proves pessimistic has a delighted customer. A firm whose estimate proves optimistic has a complaint, in this sector one that fits a familiar pattern.

What that means for the defaults. Set them low.

Deliberately conservative, then say so. A tool that states it uses cautious assumptions is making a claim about its own integrity that costs nothing and is easily verified.

What must never be used. A range presented as a point.

Where an outcome could fall across a spread, show the spread. Choosing the top of it and printing that as the answer is the same error with extra steps.

The maintenance problem nobody plans for

Tariffs And Rates Change

Any tool built on current energy prices begins ageing the day it launches. Most are built once and never revisited, which means most are quietly wrong.

What goes stale. The inputs, not the logic.

The arithmetic stays correct. The electricity price, the export rate and the assumptions about what a household pays all move. The output moves with them without anybody noticing.

Why nobody catches it. There is no error.

A broken tool announces itself. A tool producing plausible answers from outdated prices looks exactly like a working one, so it can run wrong for years.

What it needs. A date and an owner.

A stated review date, a named person responsible and a schedule more frequent than annual while energy pricing moves as it currently does.

The rule that follows. Out of date is worse than absent.

A firm that cannot commit to maintaining a calculator should not publish one, because a stale tool produces confident wrong numbers rather than no numbers.

The technical rule, a standing position

Never Mark It Up

Structured data is an assertion. Marking up a saving or a payback period asserts it as fact in machine readable form, which is the one thing the whole of this page argues against.

Why markup is different from prose. Qualification does not survive.

A page can say a figure is illustrative and based on assumptions. A structured data field carries the number and nothing else, so every qualification block three insisted on is stripped away.

What that produces. A bare claim, repeated.

The figure can then be surfaced elsewhere without any of its context. A firm has published an unqualified financial claim without ever writing an unqualified sentence.

What is barred. Any monetary or output value.

No saving, no payback period, no generation figure and no export rate in any structured field. This is the standing position across our programme rather than a rule invented for solar.

Where the mechanism is explained. Its own cluster.

Our schema material covers what structured data does and why an assertion in markup carries the weight it does.

Straightforward about what it is for

It Is A Data Capture Tool As Well

Most calculators exist to collect contact details. Customers know this perfectly well. Being straightforward about it converts better than pretending otherwise.

Why pretending fails. The reader has met one before.

Anybody who has used a few of these knows the pattern. A tool that behaves as though the exchange is a secret is treating a well informed reader as though they were not.

The rule about sequencing. Do not switch it.

Never require details before showing a result if the page implied a result was coming first. That is a bait, it is remembered as one and it is the specific behaviour this sector is judged for.

The version that works. Say what happens.

Give something useful, then explain that a fuller assessment needs a conversation and ask. Somebody who has had a straight exchange is considerably more likely to give a real number than somebody who felt caught.

What to avoid entirely. Anything with a clock.

No urgency attached to the result, no offer expiring and no scarcity. Refused across this cluster, per showing prices on a solar website.

The constructive close

What A Better Version Looks Like

Five characteristics. A tool with all five is both more defensible and more useful than the standard version.

It asks about usage rather than only about roof size. The main improvement.

Most tools ask how big the roof is, because that is easy. The more informative question is when the household actually uses electricity, which is the factor that moves the answer most.

It shows its assumptions. Per block four.

Beside the answer, in the same visual weight.

It gives a range rather than a figure. Open about uncertainty.

A spread communicates that the outcome depends on things not yet known, which is accurate and prepares the customer for a survey.

It uses conservative inputs. Per block six.

And says so, which is a credibility signal in itself.

It offers a survey as the way to get a real answer. Per block eleven.

Which is what the tool is for. The pricing boundary is in the complete guide to SEO for solar companies.

Where everything above points

The Survey Is The Only Real Answer

Every caution on this page arrives at the same place. Only an assessment of the actual property, with the household's actual consumption, produces a figure anybody should rely on.

What that makes the tool. A step rather than a destination.

The calculator's job is to get somebody interested enough to book an assessment. It was never going to produce the real number and it does not need to.

Why that reframing helps commercially. It lowers the burden.

A tool trying to be definitive has to be accurate. A tool trying to start a conversation has to be useful and straightforward, which is a much easier standard and a safer one.

What to say at the end of the output. One line.

That the only way to know is to look at the roof and the bills, then that the firm will do that. Stated as the obvious next step rather than as a sales prompt.

What we measure. Surveys booked.

Not tool completions. How we work is on our solar company SEO page and the full series is in our SEO guides for solar companies.

SEO for solar companies

Show the
assumptions.
Not just the answer.

The tool kept because it converts, its output treated as the claim it is, every assumption shown beside the figure rather than beneath it, defaults set conservatively because the two errors are not equal, with nothing marked up in structured data.

What is included every month:

Google Maps optimisation Full website management SEO campaign AI optimisation (GEO) Facebook Instagram LinkedIn Quarterly audits Monthly reporting
£350 per month, fixed

One monthly rate covering everything listed above. No setup fee. Nothing billed separately.

The full guide series

Every guide.
One practice.

The complete guide, trust signals for installers, panel installation, battery storage, showing prices on a solar website, plus presenting savings estimates responsibly.

Questions people ask

Savings Tool Content

Are you telling us to remove our calculator?
No. The tool is valuable and the instinct to have one is right, because it answers the question everybody arrives with and converts better than anything else on a solar site. The argument is about the difference between a version that helps and a version that creates a liability. That difference is mostly whether the assumptions are visible.
Why is a calculator output treated as a claim?
Because the customer relies on it. A firm producing a number thinks of it as a rough guide. The customer thinks of it as what the company told them. Where those diverge later, it is the customer's version that gets repeated. The complaints this industry is remembered for turn on people who were given figures and did not get them.
What is the single most important change we can make?
Show the assumptions beside the answer. The electricity price, the export rate, the usage pattern and the generation estimate the tool used. Somebody who can compare the assumed electricity price with their own bill either gains confidence in the number or adjusts and gets something useful. It also turns an assertion into a conditional statement.
Should the defaults be favourable to make it look attractive?
The opposite. The asymmetry settles it. A firm whose estimate proves pessimistic has a delighted customer. A firm whose estimate proves optimistic has a complaint that fits a familiar pattern in this sector. Optimism creeps in without malice, because each individual input leans slightly one way and together they produce a number nobody achieves.
Can we mark the savings up in structured data?
Never. This is a standing position across our programme rather than a solar rule. A page can say a figure is illustrative and based on assumptions. A structured field carries the number and nothing else, so every qualification is stripped away and the figure can be surfaced elsewhere without context. That is an unqualified financial claim you never wrote.
Do we need to keep it updated?
More often than annually while energy pricing moves as it does, with a stated review date and a named owner. The logic stays correct while the inputs go stale. Nothing announces the problem. A tool producing plausible answers from outdated prices looks exactly like a working one. If you cannot commit to maintaining it, do not publish it.