Marketing Agency Pricing

Updated September 2026 · Written and maintained by the Progression Agency strategy team

The complete pricing reference for buying marketing services in 2026. Every service, every model, every band — with the numbers agencies quote and the arithmetic behind them. What a retainer buys in hours. What agencies actually cost to run, so you can tell a fair margin from an unfair one. How to read a proposal, what to negotiate, and which contract clauses cost more than the price. Nothing here is gated and none of it requires talking to us.

The short answer, before the detail

Everything below is the working behind those six numbers. If you only read one section, make it ‘what agencies actually cost to run’ — understanding an agency’s own economics is the fastest way to tell a fair quote from a bad one.

Why marketing agencies do not publish prices

Four reasons. Three are legitimate. One is not, and it is worth knowing which is which.

1. The work genuinely varies. ‘Social media management’ covers a solo consultant scheduling twelve posts and a team producing forty original assets with paid amplification. A single published price would be wrong for nearly everybody.
2. Scope discovery takes a conversation. An agency that quotes before understanding your close rate, your capacity and your margins is guessing, and the guess will be wrong.
3. Competitors read pricing pages. A real commercial concern, though a smaller one than agencies believe.
4. Opacity allows price discrimination. This is the one that costs you. An agency that learns your budget before quoting can quote to your budget rather than to the work.

The nine ways agencies charge

Retainers: what should be in one

• A rolling 90-day roadmap with stated expected outcomes
• A monthly change log listing every action taken, with dates
• Reporting that leads with qualified leads and cost per acquisition including the fee
• A named person who does the work and answers email
• The right to reduce or pause when the work runs out, with 30 days’ notice

Projects: what belongs in the contract

• A written definition of done
• A fixed number of revision rounds
• Itemized deliverables with acceptance criteria
• Payment milestones tied to deliverables rather than dates
• For anything touching URLs: a redirect map before development starts

Percentage of ad spend: the conversation to have before signing

It is the industry default above roughly $10,000 in monthly media, and it scales fairly with workload. It also means the agency earns more when you spend more, regardless of whether spending more is right. That is not disqualifying — it is a thing to name openly. Ask directly: what happens when you recommend we cut spend, and is there a floor that protects your revenue when we scale down?

Performance pricing: define the lead in writing

The single most disputed model in the industry, and the disputes are always definitional. Before signing, agree in writing: a minimum call duration, a geography filter, an exclusion for existing customers, an exclusion for competitors and recruiters, and a monthly random-sample audit that either side can call. Without those five clauses, you will pay for hangups.

What every service costs: the master reference

Ranges quoted across the US market in 2026, split by provider tier. These are not our rates — they are the landscape, gathered from proposals clients have shown us and from accounts we have audited or inherited. Use them to sanity-check a quote before you receive it.

What agencies actually cost to run

This is the section that changes how you read every quote. An agency’s fee is not margin — most of it is somebody’s salary. Understanding the underlying P&L tells you when a price is fair and when it is not.

A healthy small US agency runs roughly 48% salaries, 12% contractors, 7% tools, 11% overhead, 9% sales and marketing, and 13% net profit. That 13% is normal and necessary — an agency operating at 3% cuts corners or fails, and either outcome lands on you.

What this means for a $4,000 retainer

Roughly $1,920 is salary. At a blended cost of $75 an hour for the mix of people involved, that is about 26 hours of actual human work. Add contractor time and you are buying somewhere around 28 to 32 hours a month. Whether that is good value depends entirely on who those hours belong to.

Hourly rates behind the retainer

What the person doing your work costs, before the agency’s overhead and margin. Multiply by roughly 2.5 to 3.5 to get the billing rate — that multiple covers non-billable time, benefits, tools, management, sales cost and profit.

Why offshore is cheaper, and where it is and is not appropriate

An offshore production shop runs a very different P&L: lower salaries, much higher sales and marketing cost, and a much higher margin. That is not a criticism — it is a different business model with different economics. Offshore production is genuinely good value for well-specified mechanical work: migrating 4,000 product descriptions, resizing an asset library, data entry, QA passes. It is poor value for anything requiring judgment, and templated content produced at volume is now a documented quality risk rather than a neutral activity.

Splitting a budget between media, management and creative

One of the most common questions from businesses new to paid channels, and one where the honest answer is uncomfortable: below a certain total, managed advertising stops making sense because the fee consumes too much of the media.

At $1,500 total, a $675 management fee against $825 of media is an indefensible ratio — you are paying 45% of your budget for someone to manage the other 55%, and $825 buys so few clicks in most categories that there is nothing to learn from. At $12,000 total, a $1,900 fee against $8,600 of media and $1,500 of creative is entirely reasonable, because the fee now buys real optimization against a data set large enough to optimize on.

The practical threshold sits at roughly $2,500 to $3,000 in monthly media for competitive categories. Below that, the honest recommendation is to spend the money on things that do not carry a management fee: your Google Business Profile, systematic review generation, fixing the conversion path on the site you already have, and a self-managed Local Services Ads account if you are eligible. Those produce more at that budget than any managed paid program will.

Creative is a cost line, not an afterthought

The single most common budgeting error on paid social is treating creative production as free. Meta and TikTok consume creative far faster than they consume budget — a creative that performs for six weeks on Google Ads is exhausted in eight days on TikTok. Budget 10–15% of total spend for creative production from the start, or plan for performance to decay predictably in month two while everyone wonders why.

Platform fees and the things nobody quotes

• Ad platform minimums. Programmatic and CTV typically carry $5,000–$25,000 monthly minimums before anyone will take the account.
• Marketing technology licenses. Email platforms, CRM seats, SEO tools, call tracking, heatmaps, scheduling. $200–$3,000 a month, and almost never in the agency’s quote.
• Stock assets and licensing. Music, footage, imagery, fonts, and usage rights that expire.
• Talent and usage rights. A model release for twelve months costs a fraction of one in perpetuity, and agencies routinely quote the cheap one.
• Translation and localization if you operate in more than one language.
• Legal and compliance review in regulated categories — healthcare, legal, finance.
• Your own time. Approvals, interviews, asset gathering, feedback. Budget four to eight hours a month of someone senior, or the engagement stalls.

In-house, agency, freelance or hybrid — the full cost comparison

The build-versus-buy decision is usually framed as a cost question and it is really a capability question. One in-house marketer costs about the same as a mid-sized agency retainer and gives you deep context, full attention and no channel breadth. An agency gives you breadth and specialists and no context. Neither is better; they solve different problems.

The pattern we see work most reliably for businesses between roughly $3M and $25M in revenue: one internal marketing coordinator who owns the calendar, the CRM and the agency relationship, plus an agency for the specialist channels. The coordinator does not need to be a specialist. They need to be organized, inside the business, and empowered to say ‘that report does not answer my question’.

The hidden costs of hiring in-house

• Recruitment fees of 15–25% of first-year salary if you use an agency to hire
• Three to six months of ramp before full productivity
• Benefits, payroll tax and equipment, typically 25–35% on top of salary
• Tool licenses that an agency would otherwise amortize across clients
• Management time from someone who may not know how to evaluate the work
• The replacement cost when they leave, which in marketing is frequent

The hidden costs of using an agency

• Onboarding time in month one where little visible work happens
• Your own time in approvals and feedback loops
• Context loss when your account manager changes
• Scope disputes when something obvious was not written down
• The switching cost if it does not work out, which is why ownership clauses matter
• Attention that tracks fee size across their client base

How to read a proposal you have been sent

Most marketing proposals are written to look substantial rather than to be evaluated. They open with a philosophy section, present a process diagram, list activities in the present participle — ‘optimizing’, ‘monitoring’, ‘refining’ — and arrive at a monthly number without ever stating how much work it buys. Nine questions convert any of them into something you can actually judge.

1. How many hours a month, and at what seniority? Divide the fee by the hours. Refusal to answer is itself the most useful data point in the proposal.
2. What specifically changes in the first 60 days? A competent agency has already looked at your situation and can name things before you sign.
3. What is the one number we judge this by at twelve months? If the answer involves rankings, impressions or followers, renegotiate now rather than in month nine.
4. What is explicitly out of scope? The absence of this section is where every change order in history has come from.
5. Who does the work, and how many other accounts do they carry? The person in the meeting is frequently not the person on the account.
6. What do we own, during and after? Domain, hosting, ad accounts, analytics, content, creative files, call tracking numbers, and any data.
7. What is the term, notice period and renewal mechanism?
8. What does offboarding look like? Ask for it in writing before you need it.
9. What would make you tell us to stop spending? The answer tells you whether you are buying advice or execution.

Negotiating: scope, not rate

Asking a competent agency to do the same work for less money produces a worse version of the same work. Asking them to do less work for less money produces a smaller but intact program. The second conversation is the productive one and almost nobody has it.

Things that cost the agency little and protect you a lot

• A 30-day notice period instead of 90
• No auto-renewal, or renewal requiring affirmative agreement
• Owner-level access to every account from day one
• Work product transferring to you on final payment
• An open-source, self-hostable platform rather than a proprietary one
• Payment milestones tied to deliverables rather than to dates
• Category exclusivity within your actual service area
• A capped or index-linked price escalation clause
• A written offboarding process with a fixed timeline

Things not worth asking for

• A lower hourly rate for identical work — you get a more junior person
• More deliverables at the same price — you get thinner deliverables
• Guaranteed outcomes — nobody controls the platforms
• Free trial work — it selects for agencies with spare capacity, which is not a good sign
• Payment purely in arrears on a long project — it prices in their financing risk

Contract clauses that cost more than the price

The price is negotiated in the first meeting and the clauses are skimmed in the last one. That ordering is backwards. Two of the clauses below routinely cost businesses more than the entire annual fee.

Red flags in marketing agency pricing

1. A price quoted before any questions about your business. They are pricing a package, not your problem.
2. Guaranteed results or rankings. Nobody controls Google, Meta or TikTok.
3. Cold outreach claiming errors on your website. They ran a free tool against 40,000 domains and mail-merged the output.
4. Very low price with very high deliverable counts. Forty blog posts for $900 is unedited machine output or offshore volume work, and at scale it is a liability rather than an asset.
5. Refusal to state hours.
6. A price that moves when you mention another quote. In either direction.
7. ‘Unlimited’ anything. Nothing is unlimited; the limit is simply undisclosed.
8. Reporting that leads with impressions, reach or followers.
9. No named team. If you cannot learn who does the work, it is because the answer would cost them the deal.
10. No client willing to speak to you. Every agency has at least one happy client with a phone.
11. Links or placements priced per unit. Priced links are paid links, and the risk lands on your domain.
12. Pressure to sign this week. Discounts that expire are a sales technique, not a price.

Working out what you can actually afford

Do this before any agency quotes you. It takes ten minutes with numbers you already have, and it converts the entire conversation from a price negotiation into a return calculation — which is a conversation you are much better positioned to win.

1. Average customer value. Not the value of one job — the total value of a customer including repeat work and referrals over a realistic horizon.
2. Your close rate on inbound web leads specifically. Look it up in your CRM. Do not estimate it; almost everyone estimates it too high.
3. Revenue per lead = customer value × close rate.
4. Maximum affordable cost per lead = revenue per lead × your target marketing cost percentage. Service businesses commonly run 10–25%; ecommerce is usually lower; high-margin professional services can go higher.
5. Leads required to justify a given fee = total monthly cost (fee plus media) ÷ maximum affordable cost per lead.
6. Sanity-check against available demand. If your county produces forty relevant searches a month and you need thirty leads to break even, the arithmetic does not work at that price no matter how good the agency is.

A worked example that works

A commercial HVAC contractor. Average customer value $28,000 including service contracts. Close rate on inbound web leads 22%. Revenue per lead $6,160. At a 12% marketing cost target the ceiling is $739 per lead. A $9,000 monthly total program needs 12.2 leads a month. In a market with real commercial demand that is achievable by month nine, and the program is comfortably justifiable.

A worked example that does not

A nail salon. Average customer value $620 across a year. Close rate 71%. Revenue per lead $440. At a 12% target the ceiling is $53 per lead. A $4,000 monthly program needs 75 leads a month. In most local markets that is not available at that cost, and the honest recommendation is a much smaller local program — profile, reviews, SMS retention — plus referral mechanics, not a $4,000 retainer.

Why the same quote is right for one and wrong for the other

Both businesses could receive an identical $4,000 proposal from an identical agency doing identical work. For the first it is excellent value; for the second it is indefensible. Price alone tells you nothing. Any agency that quotes without asking for your customer value and close rate is not in a position to know whether its own proposal makes sense — and you should say so.

How much should you spend in total?

The common benchmark is 5–12% of revenue, and like most benchmarks it is a starting point rather than an answer. The variables that actually matter are your growth ambition, your margin, and how much of your revenue already arrives through channels you do not pay for.

Two adjustments almost nobody makes. First, if a large share of your revenue is repeat and referral, your acquisition budget can be smaller because your retained base is doing the work — but you should then be spending some of it on retention, which is cheaper than acquisition and almost always underfunded. Second, if you are capacity-constrained, more leads are not a benefit. A roofing company that can install nine roofs a month does not want two hundred leads; it wants nine better ones and a higher price.

How allocation should shift as you grow

Under $1M in revenue, the majority of the budget should sit on conversion and local — making sure the traffic you already get converts, and that you appear when someone nearby searches. Between $1M and $5M, paid search and organic content become the growth engines. Between $5M and $25M, paid social and creative start to earn their place. Above $25M, brand and PR finally make sense, because you now have enough distribution for awareness to convert into demand rather than evaporating.

The most common and most expensive allocation error is doing this in reverse: a business under $1M spending 16% of a small budget on brand and PR, while its contact form silently fails on mobile.

Allocation by industry

Industry changes the mix more than size does. A restaurant and an industrial B2B supplier of identical revenue should spend almost inversely: the restaurant on social, creative and local; the supplier on organic technical content and a small amount of tightly targeted paid search. Applying a generic ‘balanced’ allocation to either produces a worse result than committing to the right extreme.

Where marketing budgets actually get wasted

Across every account we have audited or inherited, the waste concentrates in the same places. None of the top three requires spending more money to fix, and all three are diagnosable for free in under an hour. They persist because none of them sells as a monthly deliverable.

1. Traffic bought before conversion was fixed — 26% of all waste

The largest category every single year. A business spends $6,000 a month sending clicks to a homepage with no phone number above the fold, no service-specific landing page, and a contact form that fails silently on mobile. Doubling the traffic doubles the waste. Fixing the conversion layer costs less, takes weeks rather than months, and multiplies the return on every channel you will ever run afterwards.

2. Paid search geography and negatives misconfigured — 19%

Google’s default location setting includes people who show ‘interest’ in your area rather than only people physically in it. Combined with broad match and no negative keyword list, a plumbing account cheerfully pays for ‘plumbing salary’, ‘plumbing school’, ‘plumbing jobs hiring’ and clicks from three states away. Changing the location setting to ‘Presence’ and building a negative list from the search terms report is an afternoon of work that typically recovers 15–30% of spend.

3. Conversion tracking measuring the wrong thing — 14%

An alarming share of accounts either have no conversion tracking or count thank-you page views that also fire for other reasons. Automated bidding then optimizes confidently and efficiently toward a signal that is not a real lead. This is the most dangerous item on the list because it makes the reporting look fine while the money leaves.

4. Content published with no search demand behind it — 12%

Publishing volume mistaken for strategy. Forty posts a month on topics nobody searches for, nobody links to, and which Google eventually classifies as scaled low-value content. Four genuinely excellent pages targeting demand that exists will outperform them on every metric including cost.

5. Demand creation at too small a scale — 11%

Brand and awareness spend works, at scale. Below a threshold — which varies by market but is usually well into five figures monthly — it produces impressions and nothing measurable. Small budgets spent on awareness are the most reliable way to conclude that marketing does not work.

6, 7 and 8 — rebuilds, duplicates and unused tools

Rebuilding sites that already converted, usually with an unmapped redirect plan that costs a year of organic progress. Duplicate and cannibalizing pages splitting ranking signals between your own URLs. And a subscription stack nobody has audited in three years. Together roughly 18% of observed waste, and all three are one afternoon of attention away from being fixed.

When each kind of spending starts paying back

The timing question matters as much as the price question, because a channel that pays back in twelve months is useless to a business that needs revenue in March — and a channel that pays back in three weeks and stops the day you pause it is not a strategy.

1. Conversion and site fixes pay back fastest of anything, because they multiply channels you are already running. Usually visible within four to eight weeks.
2. Paid search produces leads in one to three weeks, plateaus by month six, and stops the day you stop paying. It is a tap, not an asset.
3. Local and reviews move in four to ten weeks and compound quietly for years.
4. Organic content produces almost nothing for four months and then does not stop. Over twenty-four months it typically overtakes everything else.
5. Email and retention pay back in weeks and are the cheapest revenue in the business.
6. Brand and PR pay back over years, cannot be attributed cleanly, and are genuinely worth doing once the rest is solid.

The types of agency, and what each charges for

Choosing between them is mostly a question of what you need most: breadth, depth, seniority or price. A boutique agency gives you senior attention and less bench. A full-service agency gives you breadth and less seniority per dollar. A specialist gives you depth and a predictable recommendation. There is no configuration that gives you all four, and any pitch implying otherwise is a pitch.

Running an agency search without wasting three months

1. Write down the outcome, not the tactics. ‘Twelve qualified commercial inquiries a month by Q3’ is a brief. ‘We need SEO and social’ is a shopping list, and it invites everyone to sell you their favorite thing.
2. Shortlist three, not eight. Beyond three, evaluation quality collapses and everyone’s time is wasted.
3. Give every agency identical information. Including your close rate, average customer value, capacity and seasonality. Withholding it produces proposals you cannot compare.
4. Do not name a budget in the first conversation. Ask what it would take, then ask for scope at two clearly different price points.
5. Ask all nine proposal questions of all three agencies. Score the answers side by side.
6. Speak to two current clients and one former client of each. The former client tells you more.
7. Meet the person who will do the work, not only the person selling it.
8. Ask each one what they would tell you not to do. An agency with no answer is selling.
9. Read the contract before the last meeting, not during it.
10. Start with a shorter term than they propose. Three months proves more than a pitch deck.

What good looks like, month by month

Glossary of pricing and agency terms

What you should get for the money, regardless of the number

Some things cost the agency nothing and are absent from most retainers anyway. These are not premium features; they are the minimum conditions for you to be able to tell whether you are getting value. Ask for all ten and watch which ones produce hesitation.

A rolling 90-day roadmap

Not an annual plan — a rolling one, updated monthly, with each item carrying a stated expected outcome. The value is diagnostic as much as operational: if an agency cannot fill the next ninety days with work that has a plausible expected outcome, the retainer should shrink. That conversation is much easier to have when the roadmap exists than when it does not.

A change log with dates

Every action taken, dated. It takes an agency two minutes a day to maintain and it eliminates the entire category of ‘what exactly are we paying for’ disputes. It is also the single most useful artefact when something breaks, because it lets you correlate a traffic drop with a deploy in seconds.

Reporting that includes the fee in the cost per acquisition

An agency reporting a $95 cost per lead on media spend alone, while charging $3,000 a month to produce 40 leads, is reporting a number that is $75 lower than the truth. Your blended cost per lead is $170. Both numbers are useful; only one of them is the number your accountant recognizes, and it should be the one on page one.

The right to shrink

Programs have natural rhythms. There are months where there is genuinely less high-value work available, and a healthy relationship can say so. A contract that makes shrinking impossible guarantees that the months with less work get filled with activity rather than admitted to.

An honest account of what did not work

If six months of reports contain no failed experiments, either nothing was tried or nothing is being reported. Both are problems. The presence of a ‘what we tried that did not work’ section is one of the most reliable indicators of an agency worth keeping.

Why the first three months feel better than months nine to twelve

Agencies overservice new accounts. It is not cynical — it is how you prove value quickly and it is genuinely well-intentioned. But hours delivered well above hours the fee covers is unsustainable, and the correction is gradual and rarely announced. If your engagement felt excellent in month three and thin by month nine, this is usually what happened, and it is a pricing problem rather than a commitment problem. The fix is a conversation about scope, not a change of agency.

The costs that are never in the proposal

Comparing two proposals is meaningless if one includes creative production and the other does not. These are the line items that sit outside both the agency fee and the media budget, and they are almost never flagged.

Your own time is the most underestimated cost

Every engagement requires approvals, interviews, asset gathering, feedback and decisions. Budget four to eight hours a month of someone senior. Engagements that stall almost always stall here rather than at the agency’s end — the content sits waiting for a review that nobody has time for, and three weeks disappear. If you cannot commit that time, say so before signing so the scope can be built around it.

What each agency size actually suits

Structural fit matters before any judgment about a specific firm. A $2M business at a national agency is funding overhead it will never benefit from and will be assigned the most junior available team. A $60M business relying on a solo consultant has a single point of failure sitting across its entire growth engine. Neither is a comment on the quality of the people involved.

Does agency location matter?

For local market knowledge — which municipalities matter, how the map pack behaves in your county, what your buyers actually search — yes, meaningfully. For everything else, almost not at all, and a fully remote team frequently delivers metropolitan-quality work at a substantially lower rate. Hire local for local demand capture; hire on category expertise for everything else.

What should ship in the first sixty days

Ask for this list by name in your first month. A competent agency produces most of it before invoicing for month three. An agency still ‘getting up to speed’ at day sixty is charging you for a ramp that should have taken two weeks, and the pattern rarely improves.

Common pricing questions, answered directly

Why do two agencies quote $1,200 and $6,500 for the same brief?

Because it is not the same work. Convert both to hours and seniority and the gap usually explains itself: $1,200 buys eight to ten junior hours, $6,500 buys roughly forty-five senior-equivalent hours. Both can be honest prices. The question is which matches the job. See best marketing agency.

Should I pay an onboarding or setup fee?

Sometimes. Month one carries genuine front-loaded work — audit, access, baselines, tracking, mapping. A fee covering that is legitimate when it is itemized. A setup fee with no line items is a deposit dressed as a deliverable, and you should ask what it buys.

Is a twelve-month contract standard?

Common, and defensible only with an exit clause. Compounding channels genuinely need time, so commitment is not unreasonable. A twelve-month term with no termination provision protects only one party. Negotiate three months then monthly, or twelve months with 30 to 90 days’ notice.

Can I pay only for results?

You can find agencies offering it, and the model is harder than it looks. Every dispute is definitional: is a twelve-second hangup a lead? A competitor filling your form? An existing customer? Define a qualified lead in writing with a duration threshold, a geography filter and exclusions, plus a monthly random-sample audit either side can call.

How much should I spend on marketing overall?

Five to twelve percent of revenue for most established businesses; twenty to forty percent for early-stage or new-market entry; one to four percent for low-margin retail. Adjust down if a lot of your revenue is repeat and referral, and adjust down again if you are capacity-constrained — more leads are not a benefit when you cannot service them.

What is a fair agency margin?

Thirteen to fifteen percent net is normal and healthy. Below about five percent the agency will cut corners or fail, and either outcome is yours to absorb. Above about thirty percent you are usually funding a large sales operation rather than a delivery team.

Is offshore work ever the right choice?

Yes, for well-specified mechanical work: migrating thousands of product descriptions, resizing asset libraries, QA passes, data entry. It is poor value for anything requiring judgment, and templated content produced at volume is now a documented quality risk rather than a neutral activity.

How do I compare two proposals fairly?

Normalize them. Convert each to monthly hours and seniority, list the deliverables side by side, add the costs each one excludes, and compare the ownership and exit terms. Then ask both the same diagnostic question about your own business and compare the quality of the answers.

Should the website be included in the retainer?

Usually not. Websites are fixed-scope projects with a defined end; marketing is continuous. Bundling them obscures both and makes it impossible to evaluate either. Price them separately. See how much a website costs.

What if I want to pause the engagement?

You should be able to with 30 days’ notice and no penalty. Understand what pausing costs: paid stops immediately, review velocity stalls within weeks, in-flight content stops, and rankings decay slowly. A three-month pause typically costs about two months of recovery.

Do agencies charge more for competitive industries?

Yes, and legitimately. The service name is identical; the volume of work required is not. Competing in personal injury law in a major metro requires substantially more content, technical precision and authority work than competing in a low-competition category in a rural county.

Why does creative cost so much on paid social?

Because the platforms consume it faster than they consume budget. A creative that runs for six weeks on Google Ads is exhausted in eight days on TikTok. Budget ten to fifteen percent of total spend for production from the start, or watch performance decay predictably in month two.

What is the smallest budget worth spending?

For local work and reviews, roughly $1,000–$1,500 a month produces measurable results in less saturated markets. For managed paid search in competitive categories, below about $2,500 in media there is not enough click volume to optimize against. Below those thresholds, spend on the free high-leverage work instead. See marketing ideas for small business.

Should I use one agency or several specialists?

One agency is simpler to manage and produces better channel coordination. Several specialists produce deeper work per channel and push the coordination burden onto you. Most businesses under $25M do better with one good full-service agency plus one specialist for whichever channel genuinely matters most.

How do I know if I am being overcharged?

Convert the fee to hours and ask who those hours belong to. Compare against the rate card in this page. Then check the change log: if a $6,000 monthly fee produces four blog posts and a report, you are paying roughly $1,500 per blog post, and you should say so plainly.

What should reporting cost?

Nothing extra. Reporting is part of the fee, not a line item. Any agency charging separately for a dashboard that repackages Search Console and Google Ads data is charging you for something you already own.

Do I need a contract at all for a small engagement?

Yes, and a short one is fine. At minimum: scope, fee, term, notice, ownership of accounts and work product, and what happens on exit. A one-page agreement covering those six things prevents almost every dispute we have watched happen.

What happens to my data when I leave?

Whatever the contract says, which is usually nothing. Agree it before you sign: a written offboarding process, a fixed timeline, transfer of account ownership, export of creative source files, and porting of any call tracking numbers. Asking for this at the start is routine; asking after giving notice is a negotiation you will lose.

Is it cheaper to hire in-house?

Only above a certain scale, and ‘cheaper’ hides a capability difference. A loaded in-house hire costs $85,000–$150,000 and covers one or two disciplines well. Most businesses between $3M and $25M end up with an internal coordinator plus an agency. See what is a fractional CMO.

What is the most expensive mistake businesses make?

Buying traffic before fixing conversion. It is 26% of all the waste we measure, it is free to diagnose, and it compounds — every dollar spent on traffic while the conversion layer is broken is a dollar multiplied by a number smaller than one.

Watch: verifying what a retainer produced

Three from Google Search Central that bear directly on the material above.

Every guide we publish, grouped by what you are trying to do.

Amazon ads: where they fit in a budget

Answer first: Amazon ads belong in the budget of anybody selling physical products on Amazon, and nowhere else. They are a retail media channel, which means the spend competes for placement inside a store rather than for attention across the web, and the economics are governed by margin per unit rather than by cost per click.

The practical difference from search or social advertising is that the shopper is already in buying mode, so conversion rates are higher and competition for the same keywords is fiercer. Treat it as a shelf-space cost rather than as awareness spend, and judge it on advertising cost of sale against unit margin.

What each channel costs to run properly

Every channel has a floor below which it does not work — not a floor where it works less well, but one below which it produces nothing measurable. Starting under the floor is the single most common way businesses conclude that a channel ‘does not work for us’.

Local SEO and email have no floor worth speaking of, which is why they are the right starting point for almost every small business. Programmatic display and connected TV have floors in five figures, which is why a $3,000 test of either produces impressions and nothing else. Paid social sits in between and its floor is set by creative production rather than by media.

How much businesses at your size actually spend

Useful as a sanity check rather than as a target. A $6M business spending $1,200 a month is an outlier and it is worth understanding why — sometimes it is a highly referral-driven business where that is genuinely correct, and sometimes it is a business that has never measured what it is leaving on the table. A $900k business spending $12,000 a month is an outlier in the other direction and usually running out of runway.

Where more fee stops buying more outcome

Diminishing returns arrive much earlier for small businesses, for a simple structural reason: there is a finite amount of search demand and a finite number of high-value fixes available. A business under $1M in revenue typically exhausts the genuinely high-return work at around $2,500 to $3,500 a month, after which additional fee buys activity rather than outcome. A $15M business is still getting real return from the eighth additional thousand, because it has more pages, more markets, more products and more measurable levers.

What separates a $2,000 month from a $6,000 month

Not effort, and not care. At $2,000 there are roughly fourteen senior-equivalent hours, and they go on content and local work because those produce the most per hour. Outreach and conversion testing do not appear — not because they are withheld, but because there are not enough hours in the fee to begin them. At $6,000 they appear. That is the whole difference, and it is worth understanding before assuming a cheaper agency is simply less committed.

A monthly self-check

Run the ten-point check above every month. Three or more amber or red for two consecutive months is a conversation rather than a cancellation — most of these items are fixable simply by asking for them, and an agency that responds well to being asked is usually worth keeping.

Marketing quotes, and why they are not a strategy

Answer first: the best marketing quote is the one you can act on, and almost none of the famous ones qualify. Inspirational marketing quotes circulate because they are memorable rather than because they are operational, and a quote about knowing your customer does not tell you which page to write on Tuesday.

There is a legitimate use for them, which is compressing a principle you already understand into something a team can repeat. Quotes about marketing strategy work as shorthand between people who share the underlying reasoning and fail completely as a substitute for it. If a quote is doing the work of a decision, the decision has not been made.

The principle behind the best quotes about marketing

Most of the durable ones say a version of the same thing: understand who you are selling to before deciding what to say. That is genuinely the highest-leverage idea in the discipline, and it is also the one most often skipped in favor of tactics, which is presumably why it keeps being restated.

What to do if you cannot afford an agency yet

We publish this knowing it costs us retainers, because a business that has done everything below and still needs help is a far better client than one that has done none of it. There is a substantial list of high-return work that costs nothing but attention, and doing it first means that when you do hire someone, you are paying them for the hard parts rather than the obvious ones.

Week one: find out what is actually broken

• Test your own contact form from a real phone. Submit it, and confirm the email arrives. Silent form failures produce no complaints and no leads, and we find one on roughly one site in six.
• Open Google Search Console and look at the Pages report. How many of your pages are actually indexed? If the number is much lower than the number you have published, that is your first problem and no amount of new content will fix it.
• Visit yourdomain.com/robots.txt. Check what you are blocking. A security plugin may have made decisions you never agreed to.
• Search your own brand name. Look at what appears — your site, your Google profile, directories, reviews, competitors bidding on your name.
• Call your own business from an unknown number during business hours and see what happens.
• Ask your last twenty customers how they found you. Write down the answers. This is better attribution data than most paid tools produce.

Week two: claim what is free

• Complete your Google Business Profile properly. Audit the primary category against the businesses ranking above you from inside your service area. Fill in every service with a real description. Add products. Upload photographs of actual work. Seed and answer your own questions. Set attributes and holiday hours. This is the single highest-return hour of work available to any local business and most profiles are a third complete.
• Claim Apple Business Connect and Bing Places. Ten minutes each.
• Fix your name, address and phone so they are identical everywhere they appear, down to the punctuation.
• Find and remove duplicate listings.

Week three: build the review engine

Reviews are the highest-leverage local ranking factor you control and the cheapest to move. Build a process, not a campaign: a text message sent within twenty-four hours of every completed job, to every customer, asking a specific question — ‘would you mind mentioning what we did and where?’ — and send it every week without exception. Never route unhappy customers to a private form while sending happy ones to Google; that is a policy violation that has cost businesses their entire review history.

Week four: fix the obvious conversion problems

• Put a tappable phone number in the header of every page.
• Make sure the offer is legible in four seconds: what you do, where, and what to do next.
• Shorten the contact form to three fields.
• Add trust signals above the fold — license number, insurance, years in business, review count.
• Replace stock photography with photographs of your own work.
• Add a response-time promise you can actually keep.
• Put ‘starting at’ pricing somewhere, if you are in a considered-purchase category.

Month two: fix titles and internal links

In Search Console, open Performance, sort by impressions, and find pages with high impressions and a click-through rate under two percent. Those pages are ranking and being ignored. Rewrite their titles and descriptions. Then take your ten most important commercial pages and, for each, search site:yourdomain.com "topic phrase" to find every other page mentioning the topic. Add a contextual link from each of those pages to the commercial one. That is typically thirty to sixty new internal links in an afternoon, it costs nothing, and it is the single most under-used ranking lever available to small sites.

Month three: decide honestly whether you need help

If you have done all of the above and leads have improved, keep going — you may not need an agency at all this year. If you have done all of it and leads have not moved, you now have a genuinely useful conversation to have with an agency, because the easy explanations are eliminated and whatever remains is the sort of problem worth paying someone to solve. Either outcome is better than starting a retainer with a broken contact form.

Organic traffic, and paid search on Amazon

Two measurement questions that arrive when an agency’s reporting is first scrutinized.

What is organic web traffic: visits arriving from unpaid search results, as distinct from paid clicks, referrals, direct visits and social. What is organic traffic to a website matters commercially because it is the only channel that compounds — a page that answers a question keeps arriving in results for as long as it stays accurate, while advertising stops the day it stops being paid for.

Organic results in SEO have no targeting controls at all, which is the point worth carrying into any reporting conversation. A page ranks for whatever queries the engine decides it answers, and a well-written page routinely ranks for hundreds of phrasings nobody targeted. That is why organic traffic growth is measured across a portfolio of terms rather than against a keyword list.

Organic traffic tools split into two groups that are frequently confused. Search Console reports what actually happened on your own property — impressions, clicks, positions, queries — and is the only source of truth for your own site. Third-party estimators model traffic for any domain from their own click curves and are useful for competitors and useless for auditing your own numbers.

Amazon PPC is a separate discipline again, because the platform’s search engine is not Google and the buyer is already shopping. PPC advertising Amazon accounts run on sponsored product, brand and display formats priced per click, and the lever that matters most is the listing itself: a campaign pointed at a poorly written listing pays for clicks that were never going to convert.

Frequently asked questions