Attribution, cost and ROI
SEO during a recession: should you cut it?
Written by the Top SEO Experts Team
Reviewed by Ritik NamdevLast updated 19 min read
Quick answer
Do not cut SEO by reflex, and cut it only if cash, not return, is your constraint. Compare what organic leads cost per won job with the gross profit each job earns. If the cost is far below the profit, a cut saves money now and gives up margin later.
The email usually lands in the same week as a bad month: can we pause SEO until things pick up? Wrong framing. “Is SEO worth it” has no general answer.
Key takeaway Cut SEO only when cash is the problem, not return. Trim scope before you cancel. Compare cost per won job with the gross profit each job earns.
Ask which line items stop earning inside your cash horizon and which lose value if you stop. The model below reruns with your numbers.
Should you cut SEO in a recession?
Only the parts that cannot pay you back inside your cash horizon. The rest may cost less per job than your other channels, and the math below lets you check.
A cut can swap a small saving for a slow, hard-to-reverse loss. We don’t forecast recessions, and nothing here assumes one is underway.
What do the economic dates say?
The National Bureau of Economic Research (accessed 22 Oct 2026) names February 2020 (peak) and April 2020 (trough) as the most recent turning points.
Definition The National Bureau of Economic Research defines a recession as “a significant decline in economic activity that is spread across the economy and lasts more than a few months.”
Why is 2020 a poor template?
That matters because advice written during or just after 2020 makes a poor template. A lockdown that shuts whole trades is not a slower squeeze on discretionary spending.
For a local service business, the shape differs again. Whether emergency repair holds up better than deferrable projects in your market is something to read from your own lead mix. We have no sourced data that settles it.
What does the evidence say about cutting marketing in a downturn?
The best-known study found that companies that cut costs alone fared worse than those that cut selectively and kept investing. It is large-company evidence from before modern local search, so it supports a principle, not a budget.
In “Roaring Out of Recession,” Ranjay Gulati, Nitin Nohria and Franz Wohlgezogen studied 4,700 public companies through three downturns. The table shows the main results.
| Study detail | What it found |
|---|---|
| Downturns studied | 1980 to 1982, 1990 to 1991 and 2000 to 2002 |
| Did not survive a recession | 17% of the companies |
| Outperformed rivals by at least 10% on both sales and profit growth afterward | About 9% (Harvard Business Review, March 2010, reprint copy) |
The chance of that breakaway result was 21% for companies that only cut costs and 26% for those that mainly invested.
Source: Roaring Out of Recession
View the data as a table
| Item | Value |
|---|---|
| Cost cuts only | 21% |
| Mainly invested | 26% |
| Progressive | 37% |
The “progressive” group, which cut selectively and still invested in R&D, marketing and assets, had a 37% chance. It averaged 13% growth in sales and 12% in profits, against 6% and 4% for the cost-cutters.
What are the limits of the study?
The limits are plain. These were public companies from the Compustat database, not a $3 million plumbing company. The downturns ended by 2002, long before anyone searched “plumber near me” on a phone.
And the study never mentions SEO. We borrow one idea from it: selective cuts beat across-the-board cuts.
Our view Selective cuts beat across-the-board cuts. That is the one idea we borrow from a study of large public companies.
Why does cutting SEO behave differently from cutting ads?
Ads stop the day you stop paying. SEO fades slowly and rebuilds slowly, so the cost of a cut arrives late and is harder to see. That lag makes SEO look like an easy line to trim, which is exactly the trap.
What does Google say about timing?
Google says as much about timing. Google’s SEO starter guide, last updated 10 December 2025, says: “Some changes might take effect in a few hours, others could take several months.” It suggests waiting a few weeks to assess whether work helped.
Does ranking reward age?
Ranking tends to reward age. Ahrefs studied 1 million pages that its crawler first saw in September 2023.
Only 1.74% of them reached the top 10 within a year. In a separate sample of 2 million pages created in October 2023, 6.11% did. Ahrefs also found that 72.9% of top 10 pages are more than 3 years old.
Source: How long does it take to rank
View the data as a table
| Item | Value |
|---|---|
| 1M pages, first seen Sep 2023 | 1.74% |
| 2M pages, created Oct 2023 | 6.11% |
The Ahrefs page is dated 15 May 2025 and studies September and October 2023 samples. The implication for a cut: pages that rank took time to earn it.
Our view, and it is theory rather than documented fact, is that those pages do not vanish when you stop paying. They fade as competitors publish, links age and content goes stale. Nobody publishes a clean decay curve.
How is paid different?
Paid is the opposite. If you pause ads, leads drop at once and return at once.
That makes ads the easier channel to flex for cash, and SEO the one to protect where it already works. Which channel is cheaper still depends on cost per won job.
How do you run the lead-cost math?
Work out cost per won job for each channel, then compare it with the gross profit that job earns. Break-even close rate is the lowest close rate at which a channel still covers its cost: cost per lead divided by gross profit per job.
What are the five steps?
- List each marketing line and its monthly cost.
- Count the leads each line produced over the last 90 days, from call tracking or your CRM.
- Multiply by your close rate to get won jobs, then divide cost by jobs.
- Compare cost per won job with gross profit per job (revenue minus labor and materials).
- Divide cost per lead by gross profit per job to get break-even close rate, and stress it.
Our guide to call tracking covers step 2, and our post on lead value math for home services covers step 4. Count only non-branded organic leads for SEO. People who search your company name were already coming.
A plain spreadsheet does the job: one row per channel, with columns for 90-day spend, leads, qualified leads, won jobs and gross profit. Mark a lead “qualified” only if it was in your service area and wanted a job you sell. Unqualified calls pad every channel, so skipping this blurs the comparison.
A hypothetical example
All figures below are hypothetical, for an HVAC company we made up, and are not client results. Assume $1,200 gross profit per job and a 30% close rate.
Ads cost $4,000 a month at $80 per lead. SEO costs $2,500 a month and, at maturity, produces 40 organic leads.
| Measure | Paid search | SEO (mature) |
|---|---|---|
| Monthly cost | $4,000 | $2,500 |
| Leads per month | 50 ($4,000 / $80) | 40 (assumed) |
| Won jobs (30%) | 15 | 12 |
| Cost per lead | $80.00 | $62.50 |
| Cost per won job | $267 | $208 |
| Gross profit | $18,000 | $14,400 |
| Profit after channel cost | $14,000 | $11,900 |
| Break-even close rate | 6.7% | 5.2% |
Both channels clear their cost by a wide margin. Neither is a cut on return grounds.
Now stress it: close rate falls to 24% and gross profit per job to $960 after discounting. Ads then cost $333 per won job and net $627 a job. SEO costs $260 and nets $700.
For scale only, here is a paid benchmark. LOCALiQ’s home services benchmarks, last updated 23 February 2026, report median search cost per lead of $48.91 for plumbing and $101.49 for roofing and gutters.
The data covers 4,595 North American accounts from 1 May 2020 to 25 June 2021, so it reflects a pandemic window. Replace it with your own figures.
You can rerun this with your numbers in the SEO ROI calculator and the SEO cost calculator. Our post on the five numbers behind SEO ROI explains the inputs.
Want to apply this to your site?
Not sure where to start? Get a plan for your own business.
What does a six-month SEO pause cost?
In our hypothetical model, it costs more than it saves unless organic leads fade slowly. Pausing a $2,500 retainer for six months saves $15,000. That equals only 1.04 months of the $14,400 in monthly organic gross profit.
So the question becomes how fast organic leads erode. Nobody can say for your site.
The table sets out three assumptions, with leads falling by a fixed percentage each month, compounding. Lost profit is the monthly gap versus full strength, times $1,200 per job.
| Monthly erosion | Lost gross profit, 6 months | Saved | Net of the cut |
|---|---|---|---|
| 5% | $13,921 | $15,000 | +$1,079 |
| 10% | $25,675 | $15,000 | -$10,675 |
| 20% | $43,899 | $15,000 | -$28,899 |
Source: Hypothetical example
View the data as a table
| Item | Value |
|---|---|
| 5% erosion | $13,921 |
| 10% erosion | $25,675 |
| 20% erosion | $43,899 |
| Saved by the pause | $15,000 |
What if you have to restart?
That table is generous, because it assumes nothing needs rebuilding. Add a restart: after six months you resume, and leads climb linearly back to full strength over another six. Over those 12 months the net changes as shown.
| Monthly erosion | Net over 12 months, with a restart |
|---|---|
| 5% | -$8,458 |
| 10% | -$27,543 |
| 20% | -$55,462 |
We can also solve for the break-even. With the same $14,400 monthly profit, a full six-month pause only breaks even if organic leads erode by about 5.4% a month or less. Above that, the $15,000 saved is smaller than the profit lost, before any restart cost.
What does a trim change?
A trim changes the sum. Suppose a hypothetical trimmed scope costs $1,500 a month and keeps tracking and the best pages alive. It saves $6,000 over six months and breaks even if erosion stays under about 2.1% a month.
The model is simple on purpose, and the erosion rates are assumptions, not findings. You can’t know your site’s decay rate in advance, which is the case for trimming first.
Key takeaway A pause only wins if decay is very slow.
Our post on whether SEO is a one-time thing covers the maintenance side.
What should you cut, pause or keep?
Cut work that cannot reach a lead inside your cash horizon, pause expansion you can resume, and keep anything that protects leads you already get. This is our view, applied to a typical retainer for a service business.
| Workstream | Call | Why |
|---|---|---|
| Call and form tracking, Search Console and reporting basics | Keep | You cannot run this math without it |
| Technical monitoring (indexing, speed, broken pages) | Keep | Cheap, and it protects existing pages |
| Pages for your highest-margin jobs | Keep | Closest to revenue |
| Business Profile updates and review requests | Keep | Low cost, direct call volume |
| Refreshing pages that rank 4 to 15 for money terms | Keep | Cheapest path to more leads |
| New location or service-area pages | Pause | Slow payback, easy to resume |
| New blog topics and thought pieces | Pause | Long payback, weak lead link |
| Link campaigns | Pause or shrink | Resume when cash returns |
| Pages for low-margin or unwanted jobs | Cut | Leads you do not want |
| Duplicate tools, extra reports, status calls | Cut | Cost with no lead effect |
If your contract is the problem, not the scope, read our post on SEO contract terms before you cancel. A smaller monthly scope is often possible.
Google’s own guidance, last updated 5 June 2026, says to ask any SEO provider what results to expect and how success is measured. For comparing scopes, our pricing page is a place to start.
What are the decision rules?
These are our rules of thumb, not industry standards, and they work best when you ask three questions in order:
- Is the constraint demand, cash or capacity?
- What does each dollar return?
- How fast does each line lose value?
The table sums up what to do for each constraint.
| Constraint | What to do |
|---|---|
| Cash | Trim cost with no lead effect, then expansion work, then paid spend with the highest cost per won job; cancel SEO last |
| Demand | Protect lead sources you own, and refresh pages for the line still ringing |
| Capacity | Cut the channel with the highest cost per won job; extra leads are worth little |
| Cannot measure leads by channel | Fix that before cutting anything |
If cash is the constraint
Trim spend in this order:
- Cost with no lead effect
- Expansion work
- Paid spend with the highest cost per won job
Cancel SEO last, because it is the slowest to rebuild.
Set the cash horizon first. A hypothetical owner with $60,000 in the bank and $20,000 of monthly fixed costs has three months. Any workstream that needs longer than three months to produce a lead goes on the pause list, and anything that protects current leads stays.
If demand is the constraint
Leads are scarce and each one is worth more, so protect lead sources you own. Idle crews make every lead more valuable, not less.
In our view, start with your own call log. Sort 90 days of non-branded organic leads by service line.
If emergency calls hold while project quotes stall, refresh the pages for the line still ringing. Skip new pages for the line that went quiet.
If capacity is the constraint
Extra leads are worth little. A company booked three weeks out should not spend to add more of them. Cut the channel with the highest cost per won job and revisit when the calendar opens.
If you cannot measure leads by channel
Fix that before cutting anything. A cut made without tracking is a guess with a deadline. Our guide to measurement is the place to start.
What mistakes do owners make when cutting?
Most bad cuts come from reading the wrong number at the wrong time. Four versions of that mistake are common.
| Mistake | What goes wrong | What to do instead |
|---|---|---|
| Cutting in month five | A program that hasn’t produced leads yet looks like pure cost, so it gets cancelled just before the ramp | Cut scope and keep the pages already built |
| Judging SEO by branded leads | Calls from people who typed your company name inflate organic numbers | Strip them out first; if non-branded leads are thin, fix or cut the work |
| Moving money to ads without checking capacity | Paid leads arrive fast, but if crews are booked, more leads add cost and no jobs | Check the calendar first, and track paid close rate separately from organic |
| Cutting everything in one meeting | Across-the-board cuts are easy to approve and hard to reverse | Sort each line into keep, pause or cut, and write down what would trigger resuming |
Watch out Cancelling a young program in month five turns a spend into a write-off. Cut scope and keep the pages already built.
The Harvard Business Review sample suggests selective cuts did better than across-the-board cuts.
Should you start SEO in a downturn if you have not already?
Start only if your cash can carry the months before leads ramp. A delay costs time, but a start you must abandon in month six wastes the whole investment. In our own payback model, a three-month delay moved a hypothetical payback from month 25 to month 33.
Our rule of thumb is comfort with payback inside roughly 18 to 24 months, and caution beyond 30. The detailed method is in our post on the SEO payback period, and how long SEO takes covers timing. If you only need a lower-cost start, read SEO on a budget.
What the data doesn’t show
We could not find a primary source that measures how fast local rankings decay after SEO stops, so the erosion rates above are assumptions. We also have no client lead or revenue figures to publish.
- The Harvard Business Review study covers large public companies in three downturns that ended by 2002.
- The LOCALiQ benchmark covers May 2020 to June 2021, so it describes a pandemic window.
- The Ahrefs study dates from 2025 (data from 2023) and measures mostly new pages for the one-year result.
- Our model uses one gross profit per job and one close rate, and real businesses have a mix.
- Nothing here tests whether demand for your particular services falls in a downturn.
What should you do this week?
Pull 90 days of leads by channel, compute cost per won job for each, and compare it with gross profit per job. Then mark each SEO workstream keep, pause or cut, and set your cash horizon before any cancellation call.
Do the rank-4-to-15 page refreshes and tracking work first, since they are the cheapest dollars in the retainer. Revisit the model monthly for a quarter. Close rate and gross profit move, not the arithmetic.
Quick check Can you say, today, what one won job from each channel costs you? If not, fix your tracking before you cancel anything.
Frequently asked questions
Should a small business cut SEO during a recession?
Not by default. Compare what SEO costs per won job with the gross profit that job earns.
If the cost is far below the profit, cutting saves cash but gives up margin. Cut only when cash runway, not return, is the constraint, and trim scope before cancelling.
Does SEO still work when the economy slows?
People keep searching when budgets tighten, though demand for some services falls. Google says some changes take effect in hours and others take several months, so results lag spending in both directions. Judge the channel by your own non-branded organic leads and close rate, not by headlines about the economy.
What should I cut first from my SEO budget?
Start with work that has no lead path inside your cash horizon: new blog topics, pages for low-margin jobs, and report padding. Pause expansion such as new location pages. Keep tracking, technical monitoring, your highest-margin service pages and Business Profile upkeep, since those protect leads you already get.
How fast does SEO performance fall if I stop paying?
We found no sourced benchmark. Rankings can hold for a while, but competitors keep working and pages go stale.
In our hypothetical model, a 10 percent monthly erosion for six months cost more than the retainer saved. Your own Search Console trend is the real answer.
Is it smart to start SEO in a downturn?
It can be, if your cash can carry the months before leads ramp. Our rule of thumb is a payback inside 18 to 24 months. Run the numbers in an ROI calculator first, and do not start if a cash squeeze would force you to stop in month six.
Should I move SEO money to paid ads instead?
Only if paid ads have a lower cost per won job at the volume you need. Ads stop and start instantly, while SEO compounds and fades slowly. Compare both against gross profit per job, and check that your crews have capacity for the extra leads.
How do I pause SEO without losing my pages?
Leave the site live and keep hosting, the domain and basic monitoring paid. Don’t delete or redirect pages to save effort.
What you stop is new work: fresh content, link campaigns and expansion. Our view is that keeping what exists costs little and avoids a rebuild if you resume.
Run your own numbers in the SEO ROI calculator, then Get Your Organic Growth Plan.
Sources
- Business Cycle Dating, National Bureau of Economic Research, accessed 22 Oct 2026
- Roaring Out of Recession, Harvard Business Review, March 2010 (reprint copy), accessed 22 Oct 2026
- Do you need an SEO expert, Google Search Central, accessed 22 Oct 2026
- SEO starter guide, Google Search Central, accessed 22 Oct 2026
- How long does it take to rank, Ahrefs, accessed 22 Oct 2026
- Home services advertising benchmarks, LOCALiQ, accessed 22 Oct 2026
Want to apply this to your site?
Not sure where to start? Get a plan for your own business.
