Skip to main content
Contact
Mainsail vs Other small-business SEO agencies

SEO agency for small businesses: how to actually evaluate one

Most small-business SEO agency evaluations measure the wrong things. The buyer asks what the agency does; the agency pitches what they sell; both miss what actually predicts a good fit. Here's the inverse framework: what a credible agency asks YOU in discovery, how to read the answers, and the four-step verification process that filters out 80% of the bad fits before you sign.

We get asked all the time: “how do I tell a good SEO agency from a bad one?” The most common answers — check their reviews, look at their portfolio, ask about their process — are fine but miss the bigger signal. The single most predictive thing about whether an SEO agency will deliver for your small business isn’t what they pitch you. It’s what they ask you in discovery. Here’s the full framework: three discovery questions a credible agency will ask, the red flags that should end the conversation early, the portfolio test most agencies fail, and how to verify what an agency claims before you sign anything.

TL;DR

The standard advice on evaluating an SEO agency — check reviews, ask about process, look at past work — is incomplete. The signal that actually predicts good fit comes from what the agency does before the pitch.

A credible small-business SEO agency will:

  1. Ask about your business, not their service. They want to know what you sell, who buys it, and what an ideal customer looks like — before they say a word about SEO.
  2. Push back on assumptions about rankings. When you say “I want to rank #1 for [keyword],” they should ask whether ranking #1 for that keyword would actually bring you customers. The honest answer is sometimes no.
  3. Look at your current site and ask what you’d change. Not to pitch a redesign — to understand what isn’t working and why.

If an agency doesn’t ask those three things (or some version of them) in the first call, they don’t actually understand small-business SEO. They’re running a sales script.

The rest of this piece is the full evaluation framework — how to verify what they claim, what red flags should end the conversation, when to pick the cheapest option, and when to pay more.

What does a good small-business SEO agency actually deliver?

The work splits into three buckets. A credible agency does all three; a weak one does only the first.

1. On-page SEO (your website). Page titles, meta descriptions, schema markup, internal linking, page speed, mobile fit, site architecture, content. This is what most “SEO” providers actually deliver. It’s necessary but not sufficient.

2. Off-page SEO (the rest of the web). Google Business Profile management, citation consistency across 30+ directories, review acquisition and response, local press, Reddit and YouTube presence, vertical-specific directories. This is the ~70% of local SEO that lives off your website — and it’s the work most agencies skip.

3. Reporting + iteration. Tracking what’s ranking, what’s converting, what’s not, and changing course based on the data. This is what turns the first two from a project into a service.

Any agency that does only #1 is selling you a one-time setup. Any agency that does #1 + #2 is selling you a service. Any agency that does #1 + #2 + #3 is selling you a partnership.

Most small businesses with growth intent need the third. Most agencies sell the first and call it the third.

The three questions a credible agency will actually ask you

This is the single most useful inversion of the standard evaluation framework. Don’t measure the agency by what they tell you. Measure them by what they ask you.

Question 1: “Tell me about your business.”

A credible agency starts here. Open-ended, no agenda, just listening. They want to know what you sell, who buys it, where customers come from today, what an ideal customer looks like, and what’s working vs. what isn’t.

Why this matters: SEO that’s good for a Greenville restaurant is different from SEO that’s good for a Greenville HVAC company. The category structure is different, the customer journey is different, the queries are different. An agency that pitches you SEO before understanding your business is selling a template. An agency that listens for 20 minutes before saying anything technical is doing actual diagnostic work.

What good answers sound like (from the agency):

  • “Tell me about your average customer. Who is the person who’s most likely to buy from you?”
  • “What’s your biggest source of new customers today — referrals, walk-ins, ads, search?”
  • “If you could double the number of one type of customer, which type would it be?”

Red flag answers:

  • Immediately pivoting to “and what keywords do you want to rank for?” — that’s the agency template fitting your business into their service, not the other way around.

Question 2: “Do you think ranking at the top of Google will actually get you more customers?”

This is the one that separates honest agencies from order-takers. The naive answer to this question is “yes, of course — that’s why I’m hiring you.” A credible agency knows the answer is sometimes no.

The honest version: ranking #1 for a category keyword (e.g., “greenville plumber”) looks like it would bring you customers, but the data often shows it doesn’t. Category keywords are heavily searched by other agencies, tools, scrapers, and the curious — not by actual buyers in volumes that matter. Buyer-shaped queries (“plumber near me,” “emergency plumber greenville,” “best plumber for [specific problem]”) are usually a different and smaller search universe. A good SEO agency understands the difference and tells you which type of ranking is actually worth chasing for your business.

What good answers sound like (from the agency):

  • “Sometimes — depending on what people are actually searching when they’re ready to buy. Let me look at your category before I tell you what to chase.”
  • “Ranking is a lagging indicator. The leading indicator is whether you’re surfacing for buyer-intent queries, which is a different thing.”
  • “Honestly, ranking #1 for [category] in [city] might not move your call volume. Here’s why and what to chase instead.”

Red flag answers:

  • “Absolutely! We guarantee page-one rankings for [your category keyword] in 90 days.”
  • Any version of “more rankings = more customers” with no qualifier.

That kind of certainty is the agency telling you what you want to hear instead of what’s true. We wrote about exactly this gap on our own homepage — we ranked #1 organically for “Greenville SEO” and learned the query is mostly non-buyer traffic. The lesson generalizes.

Question 3: “Are you happy with your current website, or would you like anything different?”

This question reveals two things at once. First, it tells the agency where the friction is (if you say “I hate how slow it is” or “I can’t update it myself” or “I don’t know what it’s supposed to be doing”). Second, it reveals whether you’ve thought about what you actually want — which affects how the agency should scope their work.

A credible agency uses this answer to decide whether they should be selling you SEO at all. If your site is fundamentally broken (won’t load, no schema, terrible mobile experience), SEO work on top of it won’t move anything. The agency should either say “we need to fix the site first” or refer you to someone who specializes in that.

What good answers sound like (from the agency):

  • “What specifically frustrates you about it?”
  • “Have you tried updating it recently? What was that like?”
  • “Is there anything you wish it did that it doesn’t?”

Red flag answers:

  • “We’ll handle all of that as part of our SEO package.” (Either they’re not actually rebuilding the site and just claiming to, or they’re going to charge you twice for two different scopes.)
  • “The website doesn’t really matter for SEO.” (It does. A lot.)

How to read agency pricing for what it really is

Most agency pricing is engineered to be hard to compare across providers. The trick is to read three signals through the price:

1. What’s specifically named in the scope, vs. what’s vague. A scope that lists “monthly GBP posts (4), citation cleanup across the top 30 directories with quarterly re-audit, 30 target keywords tracked weekly, one strategy call per month” is specific. A scope that says “comprehensive monthly SEO program” is not. Specific scopes are easier to verify; vague scopes are easier to under-deliver against.

2. Whether the price is transparent or “request a quote.” Agencies that publish their prices have committed to a number publicly. Agencies that hide pricing calibrate it to each prospect’s budget. Both can be honest, but the publish-the-price model creates accountability that the hidden-price model doesn’t. (We wrote about why we publish ours →.)

3. Contract length and exit terms. Month-to-month with a 30-day cancel notice signals the agency expects to earn the work every month. 12-month contracts with auto-renewal signal the agency wants to lock in revenue regardless of delivery. Neither is automatically bad, but the structure tells you something about the relationship.

The pricing landscape, briefly:

  • Under $200/mo: mostly templated work. Honest if scoped that way; problematic if pitched as full local SEO.
  • $200-$400/mo: freelancer or offshore agency tier. Real work, limited scope, no AI-search workstream.
  • $400-$800/mo: lean studio tier (where Mainsail sits at $400/mo retainer). Full local SEO + AI search + monthly iteration.
  • $800-$2,000/mo: boutique agency tier. Adds content production, light paid media, vertical specialization.
  • $2,000+/mo: full agency. Multi-person team, structured account management, broad scope.

We covered the full breakdown in Affordable SEO services that actually work → and How much does local SEO cost in 2026? →

What to ask in a discovery call (a buyer’s checklist)

Inverting the previous section: here are the questions you should ask the agency, in priority order. The first three mirror what a credible agency should have asked you. The last seven are specific to evaluating fit.

  1. “What will you specifically do for me in months 1, 2, and 6?” If they can’t name the deliverables month-by-month, the scope is too vague to verify.
  2. “Who specifically will do the work on my account?” Honest answers: “I will,” “our SEO lead Sarah will,” “a team in [country] will.” Evasive answers: “our team” with no name.
  3. “Will my SOW name the specific keywords you’re targeting and the specific reports I’ll receive?” Yes is the only acceptable answer.
  4. “What’s your work on AI search optimization (ChatGPT, Perplexity, Google AI Overview)? Is it included or extra?” Some agencies haven’t built it yet; that can be fine if disclosed, dangerous if hidden.
  5. “Can I see a sample monthly report from a comparable client?” Real providers can produce this (anonymized if needed) in 10 minutes.
  6. “Where do you rank for your own primary target keyword?” If they don’t rank on page one for “[their city] SEO agency,” they can’t deliver page one for yours.
  7. “What’s your process when work isn’t producing results?” A credible answer involves data, hypothesis revision, scope change. A weak answer involves more of the same.
  8. “What’s your cancellation policy?” Month-to-month is the strongest signal. Contracts are fine if the deliverables are specific.
  9. “Can you share contact information for one current client in a similar category?” Most won’t, which is fine. The ones who do are worth talking to.
  10. “What would make you fire us as a client?” Real agencies have a profile — they know who they work with well and who they don’t. Agencies that take any client signal weak filtering.

Red flags that should end the conversation

A short checklist. Any of these on their own is a yellow flag; two or more is a walk-away.

  1. “Guaranteed page-one rankings” with no specifics on keywords or timeframes
  2. “100+ backlinks per month” packages — almost always low-trust networks that risk a Google penalty
  3. 6 or 12-month contracts with vague SOWs that say “comprehensive SEO program”
  4. No transparency on who does the work — “our team” with no names
  5. No transparency on price — only quotes after a discovery call
  6. Monthly reports as the deliverable — reports describe work, they aren’t work
  7. “Free SEO audit” sales funnels where the audit’s findings all conveniently match the standard package
  8. AI-generated content with no editorial layer — increasingly detectable, worst-performing content on the open web in 2026
  9. No mention of off-page work — schema, GBP, citations, Reddit, YouTube. On-page only = 30% of the job
  10. Confidence about AI search citations without explaining the mechanism. The honest answer is “yes, here’s specifically what we do.” The pitch answer is “we have a proprietary AI ranking method.”

How to verify what an agency claims (the four-step check)

Ten minutes of homework before any contract signing.

Step 1 (3 min): check their own SEO. Google search “[their city] SEO agency.” See where they rank. If they don’t appear on page one for their own primary target query, they can’t deliver page one for yours either. This is the single most accurate filter; most prospects skip it.

Step 2 (3 min): check their reviews. Google Business Profile reviews, mentions on Reddit, Clutch.co profile if they have one. Look for specifics. “Great communication!” is template-positive. “They moved us from page 4 to the local pack for ‘roofing contractor [city]’ in 5 months” is real.

Step 3 (2 min): check who’s writing on their blog. A blog written by ghostwriters or AI has no named authors and no specific recent posts. A blog written by humans has at least one named author, recent updates, and posts that take positions (not just generic “what is SEO” filler). Mainsail’s Journal is what this looks like.

Step 4 (2 min): ask for a comparable client reference. Even if the agency declines (most will), the way they decline tells you something. “Most of our clients prefer not to be named, but here’s a public case study” is honest. “We can’t share that information” with no offered alternative is a flag.

If they pass all four, the provider is credible at any tier. If they fail two or more, walk.

When the cheapest option is genuinely the right one

A short, honest list. Most pieces about SEO never give you this answer; it matters.

The cheap end of the market ($0-$300/mo) is the right choice when:

  • You’re pre-revenue or testing a market — pay for hosting + DIY the basics
  • Your category is low-competition in your market (rural service trades, narrow verticals)
  • Customer LTV is genuinely low — $50 customers don’t pencil out at $400/mo retainer
  • Word-of-mouth and referrals already drive 80%+ of your business
  • You have time and interest to do the work yourself

If two or more of these describe you, you don’t need our tier of agency. You need DIY + tools, or the under-$300 freelancer tier we wrote about here →.

When you should pay more

The flip side. The studio tier ($400-$800/mo) and above is the right choice when:

  • Customer LTV is meaningfully above $100 — the math compounds fast
  • You’re in a competitive category where 30% of GBP completeness loses to 100%
  • You’re trying to actively grow market share, not maintain
  • You don’t have time to manage a GBP, audit citations, respond to reviews, and keep up with the AI-search shift
  • You’d rather talk to the operator doing the work than an account manager

If three or more describe you, the studio tier earns its keep.

How to walk away from an agency gracefully

If you’ve started a conversation that’s going wrong, the exit:

  1. Decline within 24 hours of the discovery call. The longer you wait, the more sales-cycle pressure builds.
  2. Be specific but brief. “After reviewing your scope I don’t think we’re the right fit — the work I’m looking for is more weighted toward off-page than your package covers.” Honest and short.
  3. Don’t promise to reconsider later. That’s a sales-cycle invitation, not a no.

If you’re already in a contract that isn’t working, the exit:

  1. Read the contract for the actual termination clause. Most have 30-day notice; some require cause.
  2. Document the underperformance specifically. “Promised 30 target keywords tracked; we received 12 tracked.” Saves the disputed-cancellation conversation.
  3. Send written notice via email + certified mail. Creates a record.
  4. Pull your data before the agency loses access. GBP ownership, analytics access, Search Console verification, content files.

This shouldn’t be most conversations. It is sometimes.

The honest summary

The standard evaluation framework — check reviews, look at portfolios, ask about process — is fine but incomplete. The signal that actually predicts good fit is what the agency does in discovery, before the pitch.

A credible agency asks about your business, pushes back on naive assumptions about rankings, and looks at your current site honestly. A weak agency pitches a package, promises rankings, and ignores everything specific about your business.

The ten verification questions, the red flags, the four-step check before signing — those are the operational version of the framework. Run them on every provider you talk to. Most providers won’t pass them. The ones who do are worth talking to seriously.

If you want to start a conversation with Mainsail to see how we handle these questions — we’d be happy to. Even if we’re not the right fit for your specific situation, we’ll tell you that directly.

Book Email Call