Questions to Ask Before Hiring an SEO Agency

Hiring the wrong SEO agency is a quiet, expensive mistake, because the cost shows up as six months of lost time rather than a bad invoice you can dispute. This guide is for the founder or head of growth who has a shortlist and a set of sales calls booked, and who wants to walk into each one able to tell a real answer from a rehearsed one. Nearly every article on this topic hands you the questions. The harder and more useful part is judging the answers, so that is what this guide is built around: the questions worth asking, what a good answer actually sounds like, the version that should end the call, and what a competent agency delivers in the first 90 days. I answer these questions from the other side of the table, so the judgments here come from the pattern, not from a template.

The short version

• The answer matters far more than the question. A polished question list is worthless if you cannot hear the difference between a real process and confident improvisation.
• Make the agency tie the goal to revenue, pipeline, or qualified signups, not rankings and traffic. An agency that only talks positions is optimizing for the wrong number.
• Ask them to describe the first 90 days before you sign. Flexibility that cannot describe a sequence is usually improvisation wearing a nicer word.
• The named point of contact and the exit clause tell you more than the case studies do. Ask who does the work and how you leave.

What should you decide on your own side before the first call?

Four things, and settling them first changes which agency looks good. A buyer who has not defined success will be sold someone else's definition of it, usually the one that is easiest to deliver.

• The business outcome, not the metric: decide whether you are buying pipeline, qualified signups, or revenue from organic, then let the agency map rankings and traffic back to that. Walking in with "we want to rank for X" invites a proposal built around the easy keyword rather than the valuable one.
• A budget range you will say out loud: know roughly what you can commit monthly and for how long before you see return. If you are unsure what the market costs, read our breakdown of what SEO costs so the first number an agency names is not your only reference point.
• Your in-house capacity: be honest about who on your side will publish content, ship developer fixes, and approve work. An agency's speed is capped by your slowest internal bottleneck, and the good ones will ask about this early.
• A realistic timeline: organic growth compounds over months, not weeks. Our guide to how long SEO takes sets honest expectations, and an agency that promises to beat them is telling you something about its honesty.

When this does not apply: very early companies sometimes do need an agency to help define the goal itself. That is fine, as long as the first deliverable is that definition and not a content calendar.

Which questions separate a competent agency from a risky one?

The ones that force a specific process in reply. A good agency answers with method and trade-offs; a risky one answers with confidence and round numbers. The answer pattern that should worry you most is the assured figure with no mechanism behind it: a promised position, a traffic multiple, a date certain. Use the table below as your scorecard, question by question.

Ask this	A good answer sounds like	A red-flag answer sounds like
How will you set and measure goals?	Ties the work to revenue, pipeline, or signups, and proposes two or three KPIs you both agree on.	"We will get you to number one for your keywords," or talks only about rankings and traffic volume.
Who actually does the work?	Names the roles and the point of contact, and is clear about what is in-house versus subcontracted.	A vague "our team," no named contact, or a senior strategist in the pitch who vanishes after signing.
What happens in the first 90 days?	Audit first, then a prioritized roadmap, a few quick wins, and an agreed reporting cadence.	"We start publishing right away," with no diagnosis of what is actually holding the site back.
What is your approach to links?	Earned, editorial, and relevance-led through digital PR or genuine outreach, described as slow.	A guaranteed number of links per month, a private network, or anything priced per link in bulk.
What reporting will we get, and how often?	A set cadence, business metrics alongside rankings, and direct access to your own Search Console and Analytics.	A monthly PDF of keyword positions, no access to raw data, and no line connecting the work to outcomes.
What results can we expect, and when?	Ranges and dependencies, an honest "it takes months," and clarity about what is outside their control.	A specific percentage lift or ranking by a fixed date, stated as a guarantee.
How do we exit if this is not working?	A clear notice period, and you keep the content, links, and account access.	A long lock-in, or they retain the assets, the logins, or the content you paid for.

Notice the shape of every good answer: it contains a method and a limit. The red-flag column is united by the opposite, a number with no process attached. That is the single most reliable tell on these calls.

How should an agency talk about goals and KPIs?

In terms of your business, not Google's results page. The agencies worth hiring translate SEO work into the metric your board already tracks, then show how rankings and organic sessions feed it. The ones to avoid do the reverse, presenting positions and traffic as the goal and leaving you to guess whether any of it reaches revenue.

A fair question to ask directly: "If rankings improve but signups do not, what will you change?" A competent answer involves intent, conversion paths, and the pages that actually earn pipeline. A weak answer treats the ranking as the finish line. Traffic that never converts is a vanity metric with an invoice attached, and a good agency says so before you do.

What should the first 90 days actually deliver?

A diagnosis before a prescription, then momentum you can measure. When I scope a first quarter, the audit and the prioritized roadmap come before any publishing, because the alternative is paying for motion and calling it progress. Use this as a checklist in the proposal, and ask any agency that cannot describe something close to it why not.

• Weeks 1 to 4, diagnose: a technical and content audit, access to your Search Console, Analytics, and CMS, a recorded baseline of your current positions and organic traffic, and a prioritized findings list ranked by impact and effort rather than by what is easiest to bill.
• Weeks 5 to 8, prioritize and ship quick wins: a roadmap tied to the goal you agreed on, a handful of low-effort fixes already live such as title and metadata corrections, internal linking, and indexation cleanup, and a reporting cadence set in writing.
• Weeks 9 to 12, build the engine: the content and technical workstream running at a sustainable pace, the first leading indicators moving such as pages getting indexed and impressions rising on target queries, and a monthly report that shows the method behind every number, not just the number.

One honest caveat: revenue is rarely the right metric to judge the first quarter on. Indexation, crawl health, and impressions on the right pages are the early signals, and an agency that promises revenue in 90 days is either lucky or lying. When not to enforce this rigidly: a pure technical-recovery engagement can skip content entirely, and that is correct, not a shortcut.

How do you vet case studies and references without taking them at face value?

Ask for the engagements that did not work, and talk to a client who left. Anyone can assemble a reel of wins; what tells you how an agency behaves is how it describes a campaign that underperformed and what it changed in response. A studio that cannot name a single engagement that went sideways is either new or editing the truth.

• Ask to speak to a churned client, not just a happy one: the reference list is curated by definition. A former client will tell you how offboarding went and whether you would have kept the assets.
• Check that the work is actually theirs: for a named case study, confirm the agency did the SEO rather than riding a product launch or a brand's existing authority. "We grew their traffic" and "their traffic grew while we were engaged" are different claims.
• Match the case study to your situation: a win for a DR 80 enterprise tells you little about what happens for a young domain. Ask for work done on a site at your stage.

What should you ask about contracts, reporting, and the exit?

Three things decide how painful the relationship is to leave: lock-in length, who owns the assets, and how you offboard. Sort them before signing, because they are far harder to negotiate once the agency holds your content and logins. The engagement model also shapes this, and our explainer on how an SEO retainer works is worth reading before you agree to one.

• Term and notice: a short initial term or a clear notice period is a sign of confidence. Long mandatory lock-ins protect the agency, not you. Our guide to SEO contract terms covers the clauses that matter most.
• Asset ownership: confirm in writing that you own the content, the links earned, the site changes, and every account. If an agency builds your authority on domains or properties it controls, you are renting your own growth.
• Reporting access: you should have direct logins to Search Console and Analytics, not a filtered dashboard. An agency that keeps you out of your own data is managing your perception, not your SEO.
• Scope of work: get the deliverables, the cadence, and what is explicitly out of scope in the document. "SEO services" is not a scope; "two technical fixes, four content pieces, and a monthly report" is.

Which answers should end the conversation?

A few answers are not yellow flags to weigh but reasons to stop. Each one signals either dishonesty or a practice that can get your site penalized, and no price makes that worth it.

• Guaranteed number-one rankings: nobody can promise a position, because Google's ranking system is not theirs to command, and Google itself warns against anyone who promises rankings in its own guidance on hiring an SEO. A guarantee of deliverables is fine; a guarantee of positions is a tell.
• Link schemes and bought links: a private link network or a monthly quota of paid links is a direct violation of Google's spam policies, and the penalty lands on your domain, not theirs.
• Proprietary techniques they will not explain: real SEO is boring and legible. An agency that hides its method behind secrecy is usually hiding risk.
• No reporting and no data access: if they will not show you Search Console or let you verify the work, assume there is little to show.
• Goals framed only as traffic and keywords: an agency that never asks about your business is selling activity, not outcomes.

Agency, freelancer, or in-house: which fits your stage?

It depends on whether your gap is direction, execution, or ownership, and on how much you publish. An agency brings a ready team, tools, and cross-industry pattern recognition, and it starts fast. A freelancer costs less and suits a narrow, well-defined need. Building in-house gives you the deepest product knowledge and the tightest collaboration, but it is slow to hire and easy to get wrong early. Many teams combine them, running a lean internal owner alongside an agency or specialist.

If your real need is a strategy and a roadmap rather than hands on keyboards, the honest answer may not be an agency at all. Compare the models in our guide to hiring an SEO consultant before you commit to a retainer you do not need. The point of all of this is to buy the shape of help your problem actually has.

The buyers who choose well are rarely the ones with the longest question list. They are the ones who decided, before the call, what a good answer sounds like, so that confidence could not substitute for a process. Build that judgment once and it serves every vendor decision you make, not just this one. This week, before your next agency call, write down the two or three business outcomes you are actually buying and the one answer in the red-flag column that would make you walk. Then listen for it.