What happens if the agency doesn't deliver the expected results?

“Not published” means the figure does not appear on that company’s own website as of October 2026 — check for yourself before you sign anything, with them or with us.

First, Separate Three Causes

"Not delivering results" covers three quite different situations, and the response to each is different. Establishing which one you are in is the necessary first step, and it is frequently skipped.

The third is uncomfortable because it is not the agency's fault, and it is considerably more common than firms expect. Call recordings settle it quickly — if the calls arrived and went to voicemail during business hours, the problem is not the campaign.

The second is the genuinely hard case. Marketing can be competently executed and still not work: the market may be saturated, the practice area may not search, the positioning may be wrong, or the firm may be in a geography it cannot realistically win. A good agency will have raised these as risks early. One that promised results without qualification has a harder conversation to have.

What Recourse Actually Exists

Worth being blunt: in most legal marketing agreements, very little.

The typical agreement contains no performance obligation at all. It obliges the agency to perform services, not to achieve outcomes, and performing the services is usually demonstrable. Where a performance clause does exist, it is most often a service credit — additional months of the same work — rather than a refund. If the work has not been producing, more of it is a questionable remedy.

Refunds are rare and generally arise only where the agency plainly failed to perform, not where it performed without results. A dispute about whether "ongoing optimisation" happened is unwinnable in practice, which is precisely why scope written as countable deliverables matters so much at the start.

Litigation is almost never proportionate. The sums involved — a few months of fees — sit well below the cost of pursuing them, and the agency knows this.

Which leaves the notice period as the real remedy. On thirty days, your exposure to a failing engagement is thirty days. On a twelve-month term signed in January, a problem identified in March costs you nine more months. That difference is worth more than any performance clause you are likely to be offered.

The Conversation to Have

Before leaving, one properly structured conversation is worth having — in writing, so it exists afterwards.

1. State what you expected, with reference to what was said at the start. The baseline and the month-six expectation, if you established them.
2. State what has happened, specifically. Rankings, calls by channel, signed matters. Numbers rather than impressions.
3. Ask for their account of the gap. A competent agency will have a diagnosis, and it may be correct — including a diagnosis that implicates your intake or your capacity.
4. Ask what they propose to change, and by when. With a date attached.
5. Set a review point. Thirty or sixty days, with a stated test.

Two reasons this is worth the effort even when you are fairly sure you are leaving. It occasionally surfaces a real and fixable problem — missing conversion tracking, a redirect error, budget pointed at the wrong campaign. And it creates a documented record, which materially improves your position if there is any dispute about notice or handover.

Deciding to Leave

Four signals that the review conversation is not going to change anything:

1. No baseline exists. If nobody documented the starting position, the engagement was never auditable and it will not become so now.
2. The diagnosis is about you, with no supporting evidence. Intake and capacity are legitimate explanations — but the agency should be able to show the calls and what happened to them. An unevidenced assertion that the problem is your end is a deflection.
3. The proposal is more spend. Sometimes correct. But if the existing spend has not been shown to work, increasing it is a bet rather than a plan.
4. Nothing has changed in six months of reporting. Same metrics, same layout, no decisions. The engagement is being performed rather than managed, and that pattern does not usually reverse.

When you do leave, give notice in writing, ask explicitly for the handover list — site files, content, account access, campaign history — and confirm nothing is withheld pending final payment. Do that before the final invoice is settled rather than after.

What Prevents This at the Start

Almost everything that makes this situation manageable is agreed before it happens.

A documented baseline means "did it work" is a question with an answer. Countable deliverables mean "was the work done" is checkable rather than arguable. Assets in your own name mean leaving costs you a transition rather than a rebuild. A short notice period means the cost of being wrong is one month. And an agency that named the risks honestly at the start is one you can have a real conversation with when they materialise.

None of these are remedies. They are the reason you are unlikely to need one.

Attorney Pro Marketing's structure is built around the prevention list rather than around a remedy clause. Month to month with 30 days' notice, so exposure to a disappointing engagement is one month. A day 0 baseline and a day 14 report against it, so the auditability exists from the first fortnight. Deliverables stated as page counts per tier. Assets and accounts in the firm's name, so leaving is a transfer rather than a rebuild.

Call tracking with recordings from the Growth tier ($3,500) is what makes the three-causes question answerable rather than a standoff — if calls arrived and were missed, the recordings show it, and that is a problem worth knowing about regardless of who is at fault.

Related Questions