No Case Studies Yet: What to Ask an Early Agency

You are comparing two agencies. One has a results page full of logos, large percentages and quotes from happy owners. The other has an empty shelf and a note saying its first case studies are still being earned. The first feels safer. Before you let that feeling decide, ask where the first agency's numbers came from, and which of them you could check yourself.

I run the second kind. The NetWebMedia results page says it plainly: we would rather ship empty than ship fabricated. A founding client's case study goes up only once their retainer clears 90 days of real attribution, and only after the client has reviewed the numbers, the quote and the framing. Until then the shelf stays empty. This post is about how to judge an agency in that position, including mine.

What an empty shelf does and does not tell you

It tells you the agency is early, or that it has chosen not to publish numbers it cannot back up. It does not tell you the work is poor. A full shelf, for its part, proves that the agency is good at writing case studies. Whether the numbers inside them are real depends on two things the page rarely shows: what the starting point was, and how the result was measured.

There is also a quieter signal in how an early agency talks about its empty shelf. If it fills the space with vague praise, unnamed logos or results that cannot be traced to anyone, that tells you how it will report to you later. If it says what it does not have yet and explains how it will earn it, that tells you something too.

Questions that work when there is nothing on the shelf

What do you write down before any work starts?

An agency that measures properly starts with a documented baseline: traffic, conversions, pipeline, cost per customer, whatever matters to the business in question. Our results page commits to that as the first step of every engagement. Ask to see the template the agency uses. If there is no template, there will be no before, and without a before no after means very much.

How will you connect a result to your work?

Ask which data sources get stitched together and how a lift gets traced to a specific campaign. The honest answer includes the limits. Some results cannot be traced cleanly, and an agency that claims every improvement as its own is claiming too much. The page puts our version simply: no lifts claimed that cannot be traced to a campaign.

Who approves what gets published?

On ours, nothing publishes without the client reviewing the numbers, the quote and the framing. Ask any agency whether its existing case studies went through something similar, and whether you could speak to one of those clients directly. If the answer is no, weigh the case study accordingly.

What will you report when the numbers do not move?

Every engagement has flat months. Ask to see what a report looks like in one. An agency that can show you a clear, unflattering month is an agency whose good months you can believe when they come.

What is the trade?

Early agencies often offer something in exchange for the risk you take on. Ours is a founding-client program: below-market rates, in exchange for permission to publish your results once real attribution is in. That can be a fair trade if you understand both sides of it. Read the terms, including the minimum term and what happens to the rate afterwards, and do not treat any promise about results as a substitute for a written baseline.

How to read a case study you are shown

None of this means case studies are worthless. It means they need reading the way you would read any claim from someone who benefits from it. When an agency does hand you one, look for these before you look at the headline number.

• A start date and an end date, so you know how long the result took and whether it lasted.
• The starting figure, not only the change. A large percentage on a tiny base is still a tiny number.
• How the result was measured, and by whose tools.
• Anything else that changed in the business during the period, such as a new location, a price change or a busy season.
• A named client you could contact, or a clear reason why the client is anonymous.

A case study that has all five is worth a lot, whoever wrote it. One that has none of them is an advert, and should be weighed like one.

The risk you carry with an early agency

I will not pretend it is zero. With an agency that has no published case studies, you cannot watch its method working in your industry before you commit. You reduce that risk by keeping the first scope narrow and by knowing exactly how and when you can leave. Write the baseline down yourself as well, so the judgement stays yours. On our plans the terms are on the pricing page: a 90-day minimum, then month to month.

There is an upside that deserves a sentence too. An early agency has every reason to make its first clients succeed, because those clients are the case studies it does not have yet. It guarantees nothing, and I would not buy on it alone. Still, it is a fair reason to take the conversation seriously instead of dismissing it because the shelf is empty.

How to run the comparison

Put the same five questions from this post to every agency on your shortlist, full shelf or empty, and write down the answers side by side. Ask each one for a baseline template and a sample report from a flat month. The agencies with real case studies will answer easily, and so will the early ones worth hiring. The ones that struggle will struggle whatever is on their shelf.

What goes on the shelf, and when

The first case studies on our results page will show the baseline, the measurement and what changed, with the client's approval on every number. Until they exist, the most useful thing I can offer is the list of questions above. Ask them of every agency you talk to, and of us.

When a case study does appear there, the page promises it will carry the same numbers we report to the client, not a rounded-up version for marketing. Hold us to that. If a number on the shelf ever differs from the number in the client's own report, that is the one I would want you to ask about.

← Back to all articles