Website Marketing Budget: What to Spend Where (With Ratios)

Your website marketing budget should be allocated by your site’s stage first, then by ratio — not split across channels. Set the total at around 7% of revenue (a defensible range of 5–10%), then divide it between four buckets — measurement, fixing, strengthening, amplifying — in proportions that match the state your site is actually in. That’s the whole method. The rest of this page is the numbers.

Every budget template you’ve been handed does the same thing: a pie chart. 40% SEO, 30% paid, 20% content, 10% “other”. It looks decisive. It’s guesswork in a suit. Those splits assume every website is in the same condition — that a leaking three-year-old site and a humming one should spend identically. They shouldn’t. A site that can’t even measure its own AI referrals has a different first problem than one that’s already cited everywhere. Budget follows the problem, and the problem changes with the stage.

TL;DR

• Jump to the stage diagnostic — three questions, one row.
• Grab the ratio table — the Measure / Fix / Strengthen / Amplify percentages.
• Steal the three worked examples — $500K, $5M and $25M, line by line.

Why Channel-Split Budgets Fail Existing Sites

Channel-split budgets fail because they allocate against tactics instead of against the state your site is in. “40% SEO, 30% paid” tells you how to divide money between departments. It says nothing about whether your site is worth sending traffic to yet. That’s the only question that changes what you should spend.

Here’s the evidence that most owners can’t answer it: across our informal audit of 50 established sites, the median site appeared in just 8% of relevant AI answers — and most owners guessed either zero or far higher before we measured. If you can’t estimate your own citation share within an order of magnitude, you can’t possibly know which channel pie chart fits you. You’re not in a channel-allocation problem. You’re in a “what state is this site actually in” problem. The website marketing playbook sequences the fix — measure, then fix, then strengthen, then amplify — and this page puts a budget behind each stage of that sequence. The stage is the unit of allocation. Channels are just where the stage’s money happens to land.

How Much Should You Spend on Website Marketing Overall?

Spend around 7% of revenue on website marketing for an established site, with a defensible range of 5–10%. Go to 15% only when you’re deliberately buying growth and can stomach the losses that come before the returns. The 7% figure isn’t plucked from air: Gartner’s 2024 CMO Spend Survey put total marketing budgets at 7.7% of company revenue, and the SBA’s long-standing rule of thumb is 7–8% for firms under $5M. Website marketing is the digital slice of that — for a site-first business, it’s most of it.

What actually counts as budget matters more than the headline number, because “free” labour is why budgets lie. Here’s the line:

The founder-time line is the one everyone skips, and it’s the one that breaks the maths. If you build the plan on “I’ll just do the content myself” and never price your own hours, you’ll under-resource everything and wonder why nothing shipped. Put your time in at $100/hr. It’ll change what you choose to do yourself.

Q: Is 7% too much for a small business? No — 7% is the floor for a business that wants its website to earn more next year than this year, not the ceiling. The exception runs the other way: if your margins are thin (under 10%) or you’re in survival mode, drop to 5% and pour every point into the Fix stage, where the payback is fastest. 7% assumes growth is the goal. If it’s just to keep the lights on, spend less, but spend it all on fixing.

The Stage Diagnostic: Which Row Are You In?

Your stage is decided by three yes/no questions, answered in order. Stop at the first one that routes you.

1. Do you have 90 days of baseline numbers, including AI referrals and citation share? If no → you’re in Measure. You can’t allocate against a site you haven’t measured.
2. Does any of your five money pages convert below its own trailing 90-day baseline, or does the technical audit flag a crawl, render, or index fire? If yes → you’re in Fix. Something is leaking; plug it before you spend on anything else.
3. Is your citation share still under the 8% median? If yes → you’re in Strengthen. The site holds water but isn’t worth citing yet.

No question routed you? You’re in Amplify. Now, and only now, buy reach.

flowchart TD
    A[Start] --> B{90 days of baseline<br/>incl. AI referrals + citation share?}
    B -->|No| M[MEASURE<br/>one-off, then re-run this]
    B -->|Yes| C{Any money page below its<br/>own baseline, or a technical fire?}
    C -->|Yes| F[FIX]
    C -->|No| D{Citation share<br/>under the 8% median?}
    D -->|Yes| S[STRENGTHEN]
    D -->|No| G[AMPLIFY]

Run the questions honestly and you’ll land on exactly one row. To do them properly you need the numbers first: the 90-minute website marketing audit gives you the baseline and the money-page check, the technical SEO audit surfaces the fires, and the CRO audit tells you which money pages leak. Do the audit, then pick your row.

Budget Ratios by Stage

Here’s the asset nobody else publishes: budget ratios keyed to your site’s stage, not to channels.

¹ Measure isn’t a monthly row — nobody stays in it. It’s a single month-one spend, capped at 5% of your annual budget, that buys your baseline (the six numbers, the 20-prompt citation audit, an analytics fix). The “100%” means: while you’re in Measure, all of that one-off goes to measurement. Then you re-run the diagnostic and land on Fix, Strengthen, or Amplify.

² Measurement is a fixed 10% in every ongoing row. It is never the first thing you cut — it’s the line that tells you whether the other 90% worked. Cut it and you’re flying blind on purpose.

³ The 10% Amplify bucket in the Fix row is owned-channel maintenance only — emailing the list you already have, keeping earned mentions warm. Zero net-new paid acquisition goes to a site that still leaks; that’s the subsidy-for-a-broken-funnel mistake the playbook warns about. Paid scales in Amplify, once the funnel is proven.

What the fixed 10% measurement line actually buys: AI traffic tracking so you can see the referrals GA4 undercounts, and citation monitoring so your citation-share number stays honest quarter to quarter. Those two are how you know when to change rows.

Three Worked Examples

Three sites, three stages, three budgets. Same 7% of revenue, wildly different splits — because the stage, not the revenue, decides the shape.

$500K Services Firm, Fix Stage — $2,900/mo

A solo operator with a three-year-old site that leaks. Founder does the thinking; freelancers do the specialist work; tools stay under $300/mo; zero paid media.

Read the bottom line twice. Half of this “budget” is founder time, not cash out the door — $1,500 leaves the bank, $1,400 is the founder’s own hours priced at $100. That’s not a trick; it’s the honest cost. Ignore the imputed half and you’ll wildly over-commit your own week. What changes the row: once all five money pages clear their leaks and the technical audit passes, this firm moves to Strengthen — Fix drops from 55% to 15%, freeing roughly $1,160/mo to point at content.

$5M SaaS, Strengthen Stage — $29,000/mo

The site holds water. One in-house marketer runs the show, content contractors supply the volume, and — critically — SEO and GEO sit on one line, not two.

Notice the single SEO + GEO line. If an agency proposal in front of you has separate “SEO retainer” and “GEO/AEO retainer” rows, that’s your scripted no: they share about 80% of their inputs, so you’re being billed twice for one job. What changes the row: when citation share beats the 8% median and the topical map is over 70% published, Fix drops to 10% and Amplify rises to 45% — the SaaS moves to Amplify.

$25M Ecommerce, Amplify Stage — $145,000/mo

The site converts, it’s cited, the content inventory is clean. Now the money goes to reach: digital PR, email, and retargeting at 45%; content maintains at 35%.

One caveat that keeps this table credible: the $65,250 Amplify bucket is the website-marketing distribution slice — brand PR, email, and retargeting that compound your owned visibility. It is not an ecommerce operator’s full performance-media or blended-CAC budget. Prospecting paid, marketplace fees, and channel scaling live outside this 7% line — that’s a separate discipline with its own budget. Don’t read ~3% of revenue on distribution as “all the paid an ecommerce site runs.” It isn’t. Build the distribution engine here: digital PR and the email moat, fed by a clean content inventory. What changes the row: a new money-page leak or a failed technical audit snaps it straight back to Fix — even a $25M site amplifying a leak is a subsidy for a broken funnel.

When to Change Rows: Stage-Shift Triggers

You change rows when you hit the exit criteria, not when the quarter turns over. Each transition has a hard test.

The regression row is the one people forget. Stages aren’t a ladder you climb once; a site in Amplify that springs a leak belongs back in Fix that week. Map the whole sequence across a year with the quarter-by-quarter sequencing template, and if you’re still deciding what to fund first, SEO vs GEO vs CRO settles the order.

Budget Mistakes That Waste the Most Money

Five mistakes waste more budget than all the others combined. Every one is an allocation error.

1. Paying separate SEO and GEO retainers — billed twice for work that’s 80% identical. One visibility line, always.
2. Cutting measurement first — the moment you’re under pressure, the 10% that proves the other 90% worked is the first thing owners kill. Backwards. It’s the last.
3. Amplifying while in Fix — buying traffic for a funnel that leaks. You pay full price to lose faster.
4. Buying tools before people — a $300 stack doesn’t write the content or fix the page. Tools are the smallest line for a reason.
5. Pricing founder time at zero — the line that makes every budget lie. Unpriced hours are why plans over-commit and nothing ships.

Copy This: The Fill-In Budget Table

Ungated, no email required. It’s the ratio table turned on its side — buckets as rows, so you can fill in a single dollar column for your stage. Drop your annual revenue in, take 7% (or your chosen 5–10%), then apply your row’s percentages.

WEBSITE MARKETING BUDGET — [your site]
Annual revenue:            $__________
Budget rate (5–10%):        ____%   → Annual budget: $__________  → Monthly: $__________

Your stage (from the diagnostic): [ Measure / Fix / Strengthen / Amplify ]

MONTHLY SPLIT           Fix row   Strengthen row   Amplify row   → Your $/mo
  Measurement            10%          10%             10%          $________
  Fix                    55%          15%             10%          $________
  Strengthen             25%          55%             35%          $________
  Amplify                10%*         20%             45%          $________
                        -----        -----           -----
  Total                  100%         100%            100%         $________

* Fix-row Amplify = owned channels only (email your existing list). Zero paid acquisition.
  Measure stage = one-off, month one, capped at 5% of the annual budget, then re-run the diagnostic.

FAQ

How much does it cost to market a website? Budget around 7% of revenue as a working default, with a defensible range of 5–10%. On $500K revenue that’s roughly $2,900 a month; on $5M, about $29,000; on $25M, about $145,000. The split matters more than the total — allocate by your site’s stage (Measure, Fix, Strengthen, Amplify), not by channel.

What percentage of revenue should go to website marketing? 7% is the working default for an established site, with a defensible range of 5–10%. Gartner puts total marketing at 7.7% of revenue; the SBA’s rule of thumb is 7–8% for firms under $5M. Push to 15% only when you’re deliberately buying growth and can fund the losses that come first.

Should SEO and GEO be separate budget lines? No. SEO and GEO share about 80% of their inputs — the same content, the same crawlable site, the same entity work. Two retainers means paying twice for one job. Keep them as a single visibility line and fold the 20% that differs (entity clarity, chunk-level answers) into work you’re already doing.

Is paid advertising worth it for an existing website? Yes, if the site has passed the Fix stage. Paid multiplies a funnel that already converts and retargets warm traffic you’ve earned. On a leaking site it’s a subsidy for a broken funnel — you pay full price to lose faster. Fix your five money pages first, then scale paid.

Don’t pick a row from memory — you’ll pick the flattering one. Run the 90-minute website marketing audit first, get your six numbers and your citation share, then let the diagnostic tell you which row you’re in. Adopt it, copy the table above, and spend against the site you actually have.

Written by Sunny Patel — in SEO since 2010, specialising in semantic SEO, topical authority and AI search. Last updated 23 July 2026.

Last updated: 23 July 2026