Journal · Buying the work
An SEO ROI calculator, and why month three looks like a loss
Ad spend stops the day you stop paying for it. Search, content and operating work pay back slowly, and then keep paying. Here is the arithmetic for both, running in this page.
An owner looking at a proposal for search work, a consulting engagement or a system build is being asked to do something uncomfortable. The invoice is monthly and certain. The return is later and uncertain. Nobody can tell you the date the two cross, so most proposals do not try, and the owner is left to decide on feel.
That is a bad way to spend money, and it is avoidable. The crossing date is arithmetic. You need five numbers you already have and one honest assumption about how long the work takes to bite. This piece gives you the arithmetic, a calculator to run it in, and the reason the answer looks so different for advertising than it does for anything that accumulates.
The three shapes of a spend
Almost every line in a marketing or operations budget has one of three shapes, and the shape matters more than the size.
Advertising switches on and off. Ad spend buys attention today and stops buying it the moment the card is declined. That is not a criticism: it is the product, and it is why advertising roi is the easiest of the three to calculate and the most honest about itself. Measuring ppc roi or google ads roi is a division problem. Revenue attributable to the channel, divided by spend into the channel. Clean, fast, and gone when you stop.
Search and content accumulate. A page that ranks keeps arriving in results after the month you paid for it. So does a page that loads in half the time, a product feed that is no longer wrong, and an internal link that sends authority where it earns money. Digital marketing roi and content marketing roi on this kind of work are genuinely hard to read early, because for the first several months you are paying for an asset that has not started producing. Then the installed base starts compounding, and the arithmetic turns.
Operating work recovers margin you already earned. Fixing the records, reconciling two systems to each other, removing re-keying that somebody is paid to do: none of it brings a single new customer. It returns a few points of margin on the volume you already have, every month, forever, and it is usually the cheapest money in the building. The roi of automation and the erp roi question both live here.
Run the same monthly fee through all three shapes and you get three completely different answers about when you are ahead. That is the whole point of the instrument below, and it is worth knowing which shape you are buying before you read what we actually do or anybody else's proposal.
The calculator
Pick the shape, put your own figures in, and drag the horizon. Your figures never leave your browser and there is nothing to submit. The preset menu carries our published starting prices so you can run the arithmetic on the actual thing you would be buying rather than on a round number.
It is a marketing roi calculator, a roas calculator, a cac calculator and a payback period calculator at the same time, which is not a feature list. Those four are one piece of arithmetic asked from four directions, and splitting them across four tools is how a proposal manages to answer three of them and quietly skip the fourth.
Show the arithmetic
Show it month by month
| Month | Cost | Cumulative cost | Gross profit | Cumulative | Net position |
|---|
What each number in it means
These are the terms a proposal will use at you, with the arithmetic rather than the atmosphere. If a firm cannot produce the right hand column for its own work, that is the finding. How to calculate roas and how to calculate customer acquisition cost are the two most searched questions in this list, and each is a single line of division once somebody has decided what counts as attributable. The deciding is the hard part, and it is where the honesty goes in or does not.
Watch for one word in particular. Two of the rows below get called a payback period in ordinary speech and they measure different things: one is how long a single customer takes to repay what winning them cost, the other is how long the whole engagement takes to repay everything you have put into it. A business can be comfortable on the first and insolvent waiting for the second. If a proposal quotes you a payback period without saying which one it has calculated, it has told you less than it appears to.
| Figure | How it is calculated | What it tells you |
|---|---|---|
| ROAS | Revenue attributable to the spend, divided by the spend | A gross number. A 4x return on ad spend at 20 percent margin is losing money, which is why roas on its own decides nothing |
| CAC | Total cost of the work, divided by customers it won | Customer acquisition cost. Blended cac uses every channel and every fee together, which is the version an owner should actually watch |
| Cost per lead | Spend divided by enquiries | Only useful next to the rate at which enquiries close. Cost per acquisition is the number that pays wages |
| LTV | Average order value, times gross margin, times purchases per customer | Customer lifetime value: what one customer is worth in profit, not revenue. An ltv calculator that returns revenue is flattering you |
| LTV to CAC | LTV divided by CAC | The ltv cac ratio most operators aim at is 3 or better. Under 1 you are buying customers at a loss |
| CAC payback period | CAC divided by the gross profit on one purchase, rounded up to a whole purchase, then placed on your repeat interval to get months | How long a customer takes to repay what it cost to win them. This is the cash flow question and it is not the ROI question. You will also see it defined as CAC divided by a monthly gross profit, which is the same idea and only right if your customers buy every month. Most do not, so this page counts purchases |
| Payback month | The first month where cumulative profit exceeds cumulative cost | The crossing date. The single most useful figure in any proposal, and the one most often missing |
| Incremental revenue | Revenue with the work, minus a defensible estimate of revenue without it | The hard one. Marketing attribution exists because this subtraction is genuinely difficult, not because anyone enjoys it |
One of those rows is worth watching while you drag the horizon. The LTV to CAC tile reads 2.62x on the figures this page opens with, and it does not move, because it is computed at maturity: $1,440 of lifetime gross profit against the $550 it costs to win a customer once the work is working. The blended figure beside it does move, from 2.07x at twelve months to 2.44x at thirty six, and nothing about the business changed in between. All that happened is that the cost of the ramp got spread over more customers. A ratio that improves when you lengthen the measurement window is measuring the window. If a proposal quotes you one, ask which of the two it is.
And 2.62x is below the 3 that the table above calls the usual target. We have left it there. Tuning the defaults until the instrument cleared its own benchmark would have taken about a minute and it is the precise thing this page is arguing against. Two point something is where a great many working businesses actually sit, and what it means is specific rather than vague: the work pays for itself, and there is no room in it for the fee to rise or the close rate to fall. The three levers are your margin, the number of times a customer comes back, and your estimate of customers a month. Move any one of them and watch that tile rather than the return percentage.
Marketing roi, written plainly, is one division: the profit a spend produced, over what the spend cost. Everything difficult about it sits inside the word produced. Return on marketing investment is that same division under a longer name, and a report using the long form is not necessarily doing more work than one using the short one. Two warnings about measuring marketing roi that cost real money. First, roi tracking that credits every sale to the last click will tell you to cut exactly the work that fills the top of your funnel. Second, a marketing kpi dashboard that nobody can trace back to a record is worse than no dashboard, because it produces confident wrong decisions instead of cautious ones. Every figure in a report we send carries the query or the record that produced it, for that reason and no other.
Why month three looks like a loss
Start with the month in the title. On the figures this page opens with you are -$7,020 down at the end of month three, you have seen almost nothing back, and the invoice has arrived three times. That is the month the phone call happens. It is also the month at which the least information is available, because month three is not the bottom and is not close to it: it is about two thirds of the way down.
Watch the cumulative line instead. It goes down, keeps going down, flattens, and only then turns up. The low point is month 8, at -$10,560, and break even does not arrive until month 16. So the honest reading of month three is that it is a loss, it is supposed to be a loss, and it is going to get worse before it gets better. Nothing has gone wrong at the low point. The shape has a name, the J curve, and it is what the arithmetic of an accumulating asset looks like from inside. It is also why month three feels like a mistake to everybody who has not seen it before.
This is the honest answer to how long does seo take, and the reason the question is asked so often and answered so badly. A seo timeline is not a schedule of rankings, because nobody controls rankings. It is a schedule of causes: the technical faults fixed, the pages published, the links earned, the catalogue made accurate. Those have dates. What they produce has a lag, and the lag is the thing an owner has to be able to survive.
So does seo work, and is seo worth it? Both questions have the same answer, and it is not a yes. They are worth it if your payback month falls inside the horizon you can actually fund, and not otherwise. An owner asking is seo worth the money who cannot fund the run up to their own payback month is asking the wrong question: the right one is what else that budget could buy with a shorter payback, and sometimes the honest answer is advertising, or nothing.
Here is the part that decides it. Stop paying for ads and the revenue stops that week. Stop paying for accumulated work and it keeps producing while it decays, for months and sometimes years. The calculator reports that figure separately, because it is the entire argument for a long term seo strategy and it never appears in a monthly report.
The number nobody adds up
Every proposal is compared against doing nothing, and doing nothing is priced at zero. It never is. Doing nothing has a running cost, and most of it hides in the questions you stopped asking because the answer took three days to assemble.
Before we quote anything we count what the engagement switches off: duplicate systems, subscriptions nobody opens, card fees at the register, a cutover already quoted twice, hours of re-keying somebody is paid for. On an established business that count is often most of the fee and occasionally all of it. We do the count first and we show the arithmetic, because a price you have to find new money for is a different decision from a price that comes out of a budget you already have.
That count is the cost of inaction, and it belongs on the other side of the ledger with the rest of the total cost of ownership. The third mode of the calculator exists to put it there. Give it your annual revenue and the margin you believe is leaking, and it will tell you what recovering it is worth a month. It is the one figure on this page that requires no new customer, no ranking and no luck, which is why it usually pays back sooner than the other two: on the figures the page opens with, month 10 at +22 percent over two years, against month 16 for the search case.
One warning about that mode, and it is against our own interest to give it. The leakage figure is the input it is most sensitive to, and it is the one input we cannot check. We have put a placeholder of 1 in that field, chosen so that it does not by itself pay the fee, because a calculator on a consultant's website that arrives pre-loaded with a generous assumption about how much money you are losing is a sales instrument wearing a lab coat. We have no measured leakage figure for any business and could not lawfully borrow one from a client to put here. Replace it with yours, and if you cannot evidence yours, that is itself the finding: you are being asked to buy a recovery you cannot yet size.
What a fee buys, and what it cannot
We publish our starting prices, which makes the next part easy to check. Search and content care starts at $3,300 a month at the Essential level, $4,900 at Growth, and $7,400 at the level that includes the writing. Platform care starts at $4,400. Platform care and the Growth level together, the pairing most businesses want, are $9,300, which is those two added together and not a discount, because we would rather say so than invent one. A one-time site and page optimization starts at $5,900. Every one of those figures is on the pricing page with what it includes and what it does not.
Two of those lines, platform care and executive intelligence, reach systems that hold money and customer records, and they do not start from a price. They start after a scoping call and signed agreements covering the work, the data and confidentiality, and we will tell you plainly what we are and are not insured for before you decide anything. The calculator above will quote you a payback month on work we would not begin without those in place. That is a property of calculators, not a statement about what we are ready to sell you.
That one-time line is the website roi question in its cleanest form, and worth running through the calculator on its own: the cost lands once, in month one, the benefit is a conversion rate or a load time that improved, and the payback month is the only honest test of whether it was worth doing. Put it in the one time field and set the monthly to zero. Do that and the calculator starts fading the result from month one, because with nothing being spent nothing is maintaining it, and a fix nobody maintains is worth nothing eventually. Then do the part everyone skips: set the new customers a month to what that single job alone wins, which is a far smaller number than a retainer's. If the answer that comes back is a return you would not say out loud in front of the person paying, the figure to argue with is your customer estimate, not the arithmetic.
We are not the cheapest, and the reason is in the calculator rather than in our opinion of ourselves. A seo retainer or a marketing retainer too small to change anything still spends your money every month and still occupies your attention, and it never reaches a payback month at all. That is the worst outcome available, and it is the most commonly purchased one. Each of our starting prices is the smallest budget at which that line can move a number you would notice. If your situation needs less, we will say so on the first call and point you somewhere cheaper. That has happened.
Owners usually arrive at how much does seo cost and how much should i spend on marketing from the wrong end, which is a percentage. Marketing budget percentage of revenue benchmarks are averages of other people's businesses and they cannot know your margin, your repeat rate or your payback tolerance. Put your own five numbers in the calculator instead. A seo budget that survives contact with your own arithmetic is worth more than one that matches a benchmark.
The agency vs in house marketing question resolves the same way. A retainer here is a fraction of a part-time hire and carries no recruitment, no payroll, no notice period and no single point of failure when that person takes a holiday. The honest trade is that an employee gives you their whole attention and we give you a defined scope. If you have enough work to fill a role, hire the role, and we will tell you so.
Why the length of the engagement is the lever
Drag the horizon from twelve months to thirty six without changing anything else. On the figures this page opens with, a $3,300 a month retainer returns -19 percent at twelve months, +47 percent at twenty four and +85 percent at thirty six, finishing $101,520 ahead. The monthly fee is identical in all three. Nothing was added except time.
That gap is not a sales technique, it is the arithmetic of an accumulating asset, and it has an uncomfortable implication for how this work is usually bought. A twelve month engagement that is cancelled at month eight captures the entire cost and almost none of the return. The firm still got paid. The owner got the J and none of the curve. Consulting roi and management consulting fees get their reputation from exactly this pattern, repeated across an industry that is paid for activity rather than for accumulation.
Which is why the only version of this we are willing to sell is the long one, with nothing in it that makes leaving expensive. Thirty days' notice on any retainer, no exit fee, no clause that penalises you for going, and the cost of leaving written down before you sign rather than produced when you ask. We have to earn the renewal every month, and that is the correct incentive for work whose return arrives late. A cheaper price with a twelve month lock is not cheaper.
What we will not put in a calculator
Three things, all of them things you will be offered elsewhere.
A ranking promise. Nobody controls the results page. We will also tell you when a term we were targeting has an AI answer sitting above every organic result on it, because that changes what the traffic is worth and you should hear it from us rather than find it in a chart six months later.
A revenue guarantee. Not rankings, not revenue, and anyone guaranteeing either is selling you something. What we guarantee is process, in writing: acknowledgement inside business hours, delivery on the date we gave you, a restore actually performed rather than a backup merely taken, every figure traceable to its source, and a published correction when we get something wrong. Those are the things inside our control, so those are the things we stand behind.
Somebody else's results as yours. A case study is a number from another business with a different margin, a different repeat rate and a different starting position. It belongs in a brochure, not in your forecast. The calculator above has no numbers of ours in it except our published prices, and every figure it returns is built from yours.
If that sounds like less than other proposals promise, it is, deliberately. The six times we were wrong are written down and dated on this site for the same reason. An engagement whose returns arrive after a year is bought on whether the firm can be trusted in month four, and a firm that publishes its own corrections is making a checkable claim about that rather than an atmospheric one.
Common questions
How do you calculate SEO ROI?
Take the gross profit attributable to the work over a period, subtract everything the work cost over the same period including fees and internal time, and divide the result by that cost. The trap is the period: run it over three months and almost any accumulating work looks like a failure, because you have counted all of the cost and none of the asset. Run it to your payback month and past it. On the figures the calculator on this page opens with, break even is month 16 and the same spend returns +85 percent by month thirty six.
How long does SEO take to show a return?
The causes move inside the first month: technical faults fixed, pages published, a catalogue made accurate. What those causes produce lags them, and on ordinary figures the cumulative line does not turn for six to nine months and does not clear the fee for considerably longer. Anyone giving you a date for a ranking is guessing. The useful question is which month your own arithmetic says you break even, which depends on your margin and your repeat rate rather than on our opinion.
What is a good LTV to CAC ratio?
Three or better is the usual target, and the figure only means something if the LTV is measured in gross profit rather than revenue. Under one you are buying customers at a loss. Between one and two you are buying growth you cannot fund. Between two and three, which is where a great many working businesses sit and where this page's own default figures land, the work pays and there is no slack in it: a fee increase or a dip in your close rate takes it under. The ratio also moves with retention, which is why a business with repeat purchase can afford a customer acquisition cost that would bankrupt a single-sale business. Ask as well which CAC it was built on. A ratio computed on costs blended across a long period reads better than the same business measured at maturity, and the longer the period the better it reads.
What is a good ROAS?
There is no universal answer, because ROAS is a gross figure and your margin decides what it has to be. At 25 percent gross margin a 4x return on ad spend breaks exactly even before you have paid anyone to manage it. Convert to gross profit first, then compare against the alternative uses of the same money.
What is CAC payback period?
How long a new customer takes to repay what it cost to acquire them. Divide customer acquisition cost by the gross profit on a single purchase and round up: that is the number of purchases they have to make. Then place those purchases on your repeat interval to get the months, which is what the calculator on this page does. The commoner definition divides CAC by a monthly gross profit per customer instead. It is the same idea and it is only correct when customers buy every month, which most businesses' customers do not. Either way it is a cash flow figure rather than a profitability one: a business can have an excellent LTV to CAC ratio and still run out of money waiting for it. Where the cost cannot be repaid inside the purchases this model allows a customer to make, the calculator says Never rather than inventing a longer lifetime.
What happens to the return if I stop paying?
That depends entirely on which of the three shapes you bought. Advertising stops producing in the same week. Accumulated work keeps producing while it decays, for months and sometimes years, which the calculator reports separately. Operating work that recovered margin keeps that margin recovered until somebody breaks the process again.
How much should a small business spend on marketing?
Percentage of revenue benchmarks are averages of other businesses and cannot know your margin, your repeat rate or how many months of negative cash flow you can fund. Work it the other way: decide the payback month you can survive, then find the spend whose arithmetic clears it. If no spend clears it, the constraint is working capital rather than marketing.
Can you guarantee results?
Not rankings and not revenue, and anyone who guarantees those is selling you something. We guarantee process in writing: acknowledgement inside business hours, delivery on the date we gave you, a restore actually performed rather than a backup merely taken, every figure traceable to its source, and a published correction when we are wrong.
What is the ROI of an ERP or automation project?
Usually recovered margin rather than new revenue, which makes it the easiest of the three to defend and the hardest to get credit for. Count the duplicate systems it switches off, the hours of re-keying it removes, the stock write-offs it prevents and the card fees it renegotiates. On an established business that count is often most of the fee, and we do it before quoting rather than after.
Is this calculator saving my numbers anywhere?
No. The arithmetic runs entirely in your browser, your figures are never sent to us or to anybody else, and there is nothing to submit. Reload the page and they are gone. The page itself carries Google Analytics, like the rest of this site, so your visit is counted even though your numbers are not.
Run your own numbers, then come and argue with ours
A call is about thirty minutes. You say in one sentence what your system cannot tell you, we ask what you have, who decides and where the records live. If what you already run is adequate, we say so on that call and there is no second one. Our starting prices and what each one includes are published, so you can work out whether this applies to you before you speak to anybody.