Most SEO services aren’t sold on outcomes. They’re sold on comfort.
A polished dashboard. A “proprietary process.” A monthly call that sounds busy enough to justify the invoice. Meanwhile, the unsexy stuff—the real stuff—gets quietly deprioritized: technical triage, ruthless prioritization, content that actually earns attention, and a feedback loop that doesn’t collapse the second rankings wobble.
Here’s the thing: SEO can be a growth engine, but it’s not a vending machine. If someone is selling it like one, you’re not buying expertise—you’re buying dependence.
Setting expectations (so you don’t get played)
If you want a quick test for a vendor: ask how long it’ll take and what could make it take longer.
Anyone who answers with a clean number and no caveats is either inexperienced or performing. Real SEO has friction: dev cycles, indexing delays, competitive responses, brand limits, and plain old “Google did something weird this week.”
Now, this won’t apply to everyone, but most legitimate campaigns I’ve seen follow a rough arc:
– Weeks 1–4: discovery, baseline, technical cleanup, prioritization (not “wins,” more like stopping the bleeding)
– Months 2–3: content + internal linking + fixing systemic crawl/indexation issues
– Months 4–6: compounding effects—pages start pulling weight, conversions get measurable
– Beyond 6: you either build a machine, or you keep paying for “monthly SEO” forever
A good provider won’t promise you overnight dominance. They’ll promise you a process that survives reality. If you want to dive deeper into what goes into an effective SEO strategy, Learn more.
One-line truth:
SEO rewards consistency more than brilliance.
Guaranteed rankings are a red flag. Full stop.
If somebody guarantees “Page 1” and acts casual about it, they’re either lying or planning to take risks you’ll pay for later.
Because rankings aren’t a deliverable. They’re an output produced by dozens of inputs—some under your control (site structure, content, links, UX), some under Google’s control (algorithm changes, SERP layouts, competitors bidding on your brand), and some under nobody’s control (news cycles, trend shifts, user behavior changes).
Want something that can be guaranteed? Deliverables. Timelines. Communication cadence. An audit scope with exclusions. And a clear definition of “success” that isn’t just “more impressions.”
What a real SEO audit looks like (and what it doesn’t)
Some audits are basically a paid export from a crawler with a logo slapped on top. That’s not an audit—that’s software output.
A useful audit reads like a specialist briefing: clear prioritization, explicit tradeoffs, and recommendations that map to business goals. It should also say “we’re not doing X” when X is out of scope, low impact, or too risky.
Audit scope essentials (the stuff that actually matters)
You should see these components, stated plainly:
– Baseline diagnostics: current rankings, organic traffic by landing page, conversions from organic, indexation coverage, crawl stats
– Technical risk map: canonicalization, duplication patterns, internal linking structure, crawl traps, rendering issues, sitemap integrity
– Content gap analysis: what you should rank for (by intent), what you kind of rank for, and what’s missing entirely
– SERP reality check: who owns the results, what formats dominate (local pack, shopping, AI snippets), and what that means for click-through
– Prioritized backlog: impact vs effort, dependencies (dev/content/legal), and sequencing
– Validation plan: how fixes will be verified (not just “implemented”), plus what data sources are used
And yes, you want assumptions documented. If the forecast assumes publishing 12 articles/month but you can only ship 3, the forecast is fan fiction.
Forecasts & deliverables: stop accepting vibes
I’m opinionated here because I’ve watched companies burn quarters on pretty PDFs.
A forecast should be a model, not a promise. It needs inputs you can inspect and disagree with.
At minimum, demand:
– Data sources: Google Search Console, Analytics, crawl data, backlink tools, competitor sampling
– Assumptions: content velocity, dev velocity, link acquisition pace, seasonality, conversion rate stability
– Scenario ranges: conservative / expected / aggressive (one number is usually a lie)
– Milestones tied to levers: “If internal linking is fixed by date X, we expect Y pages to improve by Z”
And give me owner assignments in the deliverables. If everything is “recommended” but nothing is owned, nothing gets done.
A quick stat, because numbers keep people honest
Google has publicly stated that updates happen frequently—sometimes multiple times per day—via its ranking systems (broadly described across its Search documentation). That volatility is exactly why “set it and forget it” SEO fails.
Source: Google Search Central documentation on ranking systems and updates: https://developers.google.com/search/docs/appearance/ranking-systems-guide
No, this doesn’t mean chaos. It means your strategy needs monitoring and iteration baked in, not treated as a surprise expense.
Proposal red flags (pricing, guarantees, and the quiet gaps)
Some proposals look “premium” because they’re vague. That’s not premium. That’s a liability.
Look for these warning signs:
Pricing fog
If the cost is anchored to “hours” but not tied to outputs, expect slow-motion disappointment. I’ve seen 20-hour months disappear into “research” with nothing shippable to show for it.
Boilerplate masquerading as custom
If your site, your market, and your constraints don’t show up in the proposal, you’re buying a template.
The missing middle
The proposal talks big about strategy and results… but skips the messy middle: implementation, QA, stakeholder alignment, and iteration when something flops.
Report theater
If reporting is framed as the deliverable, run. Reports are receipts, not progress.
Questions that force measurable goals (and expose fluff fast)
Ask these. Don’t negotiate them down.
1) What’s the primary objective, and what’s the time bound?
Traffic is not a business outcome unless it converts. Nail the metric: leads, trials, purchases, booked calls, revenue per session.
2) What’s the baseline?
If they can’t tell you your current organic conversions by landing page, you’re not getting a performance partner—you’re getting a content factory.
3) How are we attributing conversions?
Last-click only? Multi-touch? Assisted? Different models change what “SEO success” looks like.
4) Which keywords are we targeting and why?
Intent filters matter. In my experience, “high volume” keywords are where budgets go to die if they don’t match funnel stage.
5) What does a bad month look like—and what do you do then?
This is the question most vendors hate. A serious answer includes pivots, diagnostics, and a clear escalation path.
6) What’s your link strategy, and what are your guardrails?
Quality criteria. Relevance standards. Risk controls. And yes, what happens if a placement looks toxic later.
Process over promises: the buyer framework that saves you money
Look, you’re not hiring a fortune teller. You’re hiring an operating system.
Ask what happens weekly. Ask what gets shipped. Ask how decisions get made when metrics conflict.
A solid SEO process has a cadence like this (give or take):
– Weekly: monitoring + fixes + content iteration + backlog grooming
– Biweekly: stakeholder sync + decision log updates (what changed and why)
– Monthly: performance review tied to goals, not keyword trophies
– Quarterly: roadmap reset based on what actually moved conversions
If they can’t show you how they work when rankings drop, they’re not ready for your market.
Building a scalable SEO plan that doesn’t collapse
Define goals like you mean it
Goals should survive scrutiny.
Not “rank for X.” Not “increase traffic.” More like: Increase qualified organic leads by 25% in two quarters, with landing-page-level tracking and agreed attribution.
Then tie those goals to levers:
– technical performance and crawl efficiency
– content coverage by intent (not just topics)
– internal linking that reflects priorities
– authority building that doesn’t invite penalties or reputational risk
And yes, someone needs to own each lever. “The agency” isn’t a person.
Roadmaps should be quarterly bets, not a 12-month fantasy
Long roadmaps are fine, but only if they’re modular. Google changes. Competitors react. Your product changes (hopefully). A roadmap has to bend without breaking.
I prefer a format that splits work into:
– Foundations: technical debt, indexation, templates, analytics integrity
– Compounding assets: topic clusters, programmatic pages (when appropriate), editorial content that earns links naturally
– Authority signals: PR, partnerships, digital relationships, selective outreach
– Conversion improvements: not CRO theater—actual landing page iteration tied to organic intent
Scalable tactics: boring, repeatable, profitable
The best SEO programs aren’t exciting. They’re disciplined.
Build templates. Build QA checklists. Build publishing workflows that don’t rely on one genius. Invest in internal linking systems and content refresh cycles. Prune what doesn’t perform. Keep the machine clean.
And if someone tries to sell you “secret sauce,” ask them what happens when the sauce stops working.
That answer tells you everything.