How to Check an SEO Agency Track Record
Verifying an agency is harder than it looks, because nearly all the evidence was produced by the agency and confidentiality genuinely limits what anybody can show you. This page says so first, then gives the checks that do work.
Most Of The Evidence Is Self-Reported
Case studies, screenshots and testimonials are all produced by the firm you are assessing. That does not make them false. It means they need reading rather than accepting. It also limits how far verification can go.
Why this is worth saying first. False confidence is worse.
A page implying that careful reading of case studies settles the question sends you away certain about something you cannot actually establish. Acknowledged uncertainty is more useful than that.
What self-reported means in practice. Selection.
Every firm shows its successes. Nobody publishes the client who left after four months, which is ordinary rather than deceitful. What you are reading is the top of a distribution whose shape you cannot see.
What can genuinely be checked. Four things.
Whether the evidence is internally coherent, their own visibility, references you select rather than accept, then how long clients stay. The rest of this page is those four.
What this means for your decision. Weight it accordingly.
Evidence of past work should be one input rather than the deciding one. The conversation and the proposal tell you more, because you can probe those.
What A Useful Case Study Contains
A case study is readable as evidence only if it contains four things. Most contain the last one and omit the first, which makes them unreadable however impressive they appear.
The starting position. The element that decides everything.
Where the site stood before work began. Without it, a figure means nothing, because doubling something small and doubling something large are entirely different achievements described identically.
What was actually done. The work itself.
What the agency produced or changed, rather than a summary of the approach. If a case study describes a philosophy rather than activity, nobody has told you what was bought.
Over what period. The missing dimension.
A change achieved over years and the same change over months are different results. Case studies frequently omit the period precisely because it flattens the impression.
What changed, with the basis stated. Compared to what.
Against the starting position, against the same period last year, else against something else entirely. An unstated comparison is a figure with the assumption removed.
The test to apply. One question.
Ask yourself whether you could describe what the business looked like before. If not, the case study is a testimonial with numbers in it.
Why Some Clients Cannot Be Named
An agency unable to name clients is not necessarily concealing anything. Several legitimate reasons prevent it. Knowing them stops you discarding a good firm for the wrong reason.
Clients who decline. The commonest reason.
Plenty of businesses would rather competitors did not know who does their marketing. That is their decision to make and the agency has to respect it.
Sensitive sectors. Where discretion is standard.
Some fields treat supplier relationships as confidential as a matter of course. An agency working in those will have less to show regardless of how well it performs.
White label arrangements. Naming is impossible.
Where a firm delivers work under another agency's name, it cannot identify those clients at all. That covers us, since we act as a supplier to other agencies, which is set out in what is white label SEO.
What is still reasonable to expect. Something.
Anonymised detail, a description of the business type, else one or two clients willing to speak. Complete inability to produce anything at all is different from selective confidentiality.
How to tell them apart. Ask about the constraint.
A firm with genuine restrictions explains which ones apply and offers what it can. Vagueness about why nothing is available is the version worth noticing.
Screenshots Prove Very Little
An image of a chart going upwards is the commonest evidence in this trade and among the weakest. What matters is everything the image usually omits.
What the picture does not tell you. Three things.
Where the line started, over what period it rose and what the vertical scale represents. Without those, an upward slope is a shape rather than an achievement.
Why the period matters most. Time flatters.
Almost any site that keeps publishing will show growth over a long enough window. A chart covering several years, presented without dates, is showing you the passage of time.
The scale problem. Small numbers look dramatic.
A chart with no figures on the axis can show a rise from a handful of visits to slightly more. The shape is identical to a substantial increase and the achievement is not.
Cumulative charts. Always rise.
Any figure that adds each period to the last cannot go down, including through months where nothing happened. A rising cumulative chart is arithmetic rather than evidence.
What to ask for instead. The labels.
Ask what the axis shows, what the period is and where it started. A firm confident in the result supplies those without reluctance.
Check Their Own Site
An agency's own visibility is relevant evidence and almost nobody looks. Search the terms they would want to appear for and see what happens.
Why it is relevant. They are the specialists.
A firm selling search work operates in the one market where it can demonstrate the capability directly. What its own site achieves is the least mediated evidence available to you.
The fair counter. Client work comes first.
A busy agency may reasonably prioritise paying work over its own marketing. Many good firms grow entirely on referral. Invisibility is a question rather than a verdict.
What separates the two. Whether they have an answer.
A firm that has thought about its own priorities explains them readily. One that becomes uncomfortable, else claims not to need visibility while selling it, has told you something.
What else the site shows. How they write.
The content on an agency's own site is a sample of what they would produce for you. If it is thin, generic or clearly automated, that is what your pages will be.
The version of this that applies here. Do it to us.
A reader following this advice will apply it to this site immediately, which is the correct response. Judge whether these pages resemble what you would want on your own.
Ask For A Reference You Choose
Any agency can produce a delighted client. Speaking to one they did not hand-pick is a different exercise entirely. The willingness matters as much as the conversation.
How to ask. Name the constraint yourself.
Ask to speak to a client in a similar trade, else one who has been with them a particular length of time. Setting the parameter removes the selection without accusing anybody.
Why it works. It cannot be prepared.
A hand-picked reference has usually been asked in advance and knows what is coming. One selected by your criteria has not been through that. The difference is audible.
What to ask them. Two questions.
How long they have been with the agency, then what happened when something went wrong. The second is far more informative than any account of what went well.
What reluctance means. Not automatically bad.
Some firms genuinely cannot, for the reasons in block three. What matters is whether the refusal comes with an explanation or with a change of subject.
The reference worth most. A former client.
Somebody who has left will tell you things a current client will not. A firm willing to point you at one is unusually confident. Worth asking for even though few agree.
Look For Longevity
How long clients stay tells you more than any figure a case study can contain. It is also the thing agencies least often volunteer, which is itself informative.
Why it beats results. It cannot be selected.
A firm can choose which outcomes to show. Average client tenure is a property of the whole business rather than of a chosen example, so it resists presentation.
What it actually measures. Satisfaction over time.
Clients who stay for years have kept deciding, month after month, that the arrangement is worth continuing. That is a repeated verdict rather than a single one.
How to ask. Directly.
Ask how long their typical client has been with them and who their longest client is. Both are easy questions for a firm with an answer.
The fair qualification. Young firms.
A new agency cannot have long relationships and may still be excellent. The question is proportionate: whether clients have stayed as long as the firm has existed.
The uncomfortable version. Ask why people leave.
A firm that can describe why clients have left, without disparaging them, is being straight with you about something most would avoid.
Reviews And Third-Party Listings
Public reviews and directory entries are worth a glance and should not carry much weight. They are the weakest of the checks on this page.
What they are worth. Volume and pattern.
A firm with a substantial number of reviews accumulated over years tells you it has real clients. That is a low bar and it is not nothing.
Why they establish little. Reviews reflect the relationship.
Clients rate how they were treated, which is not the same as whether the work succeeded. A pleasant firm producing poor results reviews well.
The awareness problem. Clients cannot assess it either.
A satisfied review from somebody with no way to judge the work is a statement about communication. That is worth knowing and it is not evidence of capability.
Directory listings. Mostly paid.
Appearing in a list of recommended agencies frequently reflects a fee rather than an assessment. Treat placement as advertising unless you can establish otherwise.
What to do with them. Read the middling ones.
The moderate reviews are usually the most informative, because they describe specifics rather than expressing general satisfaction or general anger.
What We Would Show You
Applying this page to us: here is what we can provide and what we cannot, including the parts that are less impressive than a page of claims would be.
What we can show. Work and method.
What we would actually do, described specifically for your business, plus examples of the kind of work we produce. That is checkable and it is the most useful thing we have.
References we can offer. Some.
Clients willing to speak, including where you set the criteria rather than us. Not every client, for the reasons in block three.
What we cannot show. Two things.
The agencies we supply as a white label partner, because those relationships are confidential. And any client result we could not evidence, which is a rule we hold across this entire site.
Why that second one costs us. It limits the claims.
A page of striking figures would be more persuasive than this paragraph. It would also require us to publish numbers whose basis we could not fully show you, which is exactly what block two tells you to reject.
What we would suggest. Weigh the checkable.
Judge us on this site, on what we say in a conversation and on whether our answers match what we have published. The full series is on the hiring an SEO agency guide.
Apply this
page to
this site.
An agency's own content is a sample of what it would produce for you, so judge whether these pages resemble what you would want on your own. Ask us for a reference where you set the criteria, then ask how long our clients have stayed.
What we commit to in writing:
Local £350 a month fixed. National £1,550 a month. Published rather than quoted on request.
Every guide.
One practice.
What an agency is, whether to buy at all, which model suits you, how to vet one, what things cost, what should be in a contract and how to work with them afterwards.
