We hired an agency. It was a disaster. Now we cannot tell what actually broke.

Three ways agency engagements break. Same three every time.

I run this marketing review a lot. The agency failure is rarely a person being bad at their job. It is one of three structural failures that were already in place at kickoff.

Failure one: strategy-light brief, execution-heavy agency

You hired an agency because you needed someone to ship. They asked for direction. You gave them the goal. They asked for the strategy. You said "you're the agency, that is what we are paying you for." So they invented strategy in week three, and the strategy drifted every month after that. Most agency engagements break in the brief, not the execution. ThinkCap Advisors' 2026 SaaS guide is direct about it: fractional CMOs are good for strategy-light teams with senior capacity. Agencies are good for execution-heavy teams with strategy already locked. Mixing the two creates the failure mode.

Failure two: KPIs measured activity, not pipeline

The agency reported click rates, impressions, MQLs, and call rates. None of it showed up as pipeline. Madison Logic's 2024 research on B2B pipeline failure points to handoff between marketing and sales as the single biggest dropoff. If the brief never named pipeline as the deliverable, the agency ran a campaign that looked successful on every dashboard except the one that mattered. Forrester research from 2023 estimates roughly 79% of marketing leads never convert and 73% never get a sales call. That math means lead-count KPIs almost always lie at growth stage.

Failure three: stage-fit mismatch

You hired the agency that worked for your friend's company at $20M ARR. You are at $4M. Or you hired the agency that ran a great PLG campaign for a self-serve SaaS, and you sell enterprise contracts. McKinsey's 2024 B2B Pulse research shows buying committees average around 10 people across 10+ interactions, and the agency that wins at one stage is structurally wrong at another. Stage mismatch is the failure mode that looks most like incompetence and is actually a hiring error.

The vendor self-description is wrong. Yours and theirs.

The B2B buying disconnect is documented. TrustRadius's 2023 study put a number on it: vendor self-description matches the buyer's actual experience 38% of the time. That gap exists between you and your customers. It also exists between you and the agency you hired. The agency's pitch deck described one company. The agency you bought described a different one. You were not lied to. You were sold to.

Madison Logic's 2024 work on B2B campaign failure adds a second piece. Even when both sides of the contract are honest, the handoff from marketing activity to sales action is where pipeline dies. The agency is paid to generate activity. Sales is paid to close. Nobody on the contract owns the bridge. If nobody on the contract owns the bridge, the bridge is the failure mode.

Run the math. If your agency cost $15K per month and ran for nine months, that is $135K. If the failure was a strategy-light brief, a $5K Marketing Strategy Review done before kickoff would have changed the brief. If it was a stage-fit error, a $500 marketing review call would have caught it on the second reference check. Most of the spend that burned was preventable at the front, not the back.

Three exercises to run on the failed engagement before the next hire.

Exercise one: the brief-deliverable diff

Pull the SOW or brief you signed at kickoff. Print it. Now print the last five deliverables the agency shipped. Compare line by line. If the deliverables match the brief and the results were still bad, the brief was wrong. If the deliverables drifted from the brief, execution was wrong. You will know inside 20 minutes which one it is. Most founders cannot find a written brief at all, which is itself a finding.

Exercise two: the pipeline trace

Pull the last three opportunities that closed-won in your CRM. Trace the original source field. If "agency-sourced" appears on any of them, write down the path. If it appears on none of them, the agency was not generating pipeline regardless of what the monthly report said. Forrester's 2023 research suggests roughly 79% of marketing leads never convert and 73% are never touched by sales, so absence of agency-sourced closed-won is not unusual. It is the question of whether anyone ever expected it to be there.

Exercise three: the stage-fit audit

Look at the agency's case studies on their site. Sort them by ARR or company size at the time of engagement. If your ARR is below or above their published case studies by a factor of two or more, you bought a stage mismatch. The agency was not wrong. They were the wrong agency.

Agency failure is not personal. The pattern is documented.

The failure modes show up in every major B2B vendor research study from the last decade. The marketing review does not invent anything. It maps your specific engagement against a known pattern.

What you walk away with after the 60 minutes.

The marketing review call is one operator and one founder in a working session. Inside 48 hours you receive a one-page written summary with the following.

• Failure mode named. Brief, execution, or stage-fit. Backed by the specific deliverable or report that proves it.
• What was preventable. The kickoff document, KPI, or reference check that would have caught it.
• Three requirements for the next vendor. Specific clauses in the SOW or specific KPIs in the contract.
• The right shape of next hire. Fractional CMO, execution agency, or internal hire. Backed by ThinkCap Advisors' fit criteria.
• Next step. If the failure was strategy, we point you to the Marketing Strategy Review. Positioning, the positioning intervention. Paid media, the Audit. Conversion, the Review. AI visibility, the Audit.

Related pain points and marketing offers.