B2B Growth Marketing Agency

Growth marketing is one of those categories where the label does most of the selling. Everyone in it says the same three words: test, measure, iterate. Ask on a random Tuesday what is being tested this week and what got retired last week, and the answers go vague fast.

I'm Léo Le Henaff, Co-founder and CTO of Moriah, a LinkedIn marketing agency and a LinkedIn Certified Marketing Partner. I own the systems behind our clients' LinkedIn workflows, so measurement lands on my desk. Our starting position is that almost any business objective has an answer on LinkedIn once the strategy is built around it, and the work is finding which one. This page covers the part most B2B growth marketing agency pitches skip: which variables get tested, how long each test runs, which number decides whether it survives, and what you actually see in a report.

The Problem: Growth Sold as a Vocabulary

When a company tells us their last agency did not deliver, the conversation almost always turns out to be about measurement before it is about execution. Nobody wrote down what "results" meant before anything launched. Six months on, the report shows impressions up and followers up, and the person who signed the contract is still hunting for pipeline that never arrived.

That is rarely dishonesty. Impressions, reach, follower counts, engagement rates: all easy to pull, easy to chart, easy to explain in a room where half the people do not work in marketing. Pipeline is harder. You have to agree in advance which conversations count as qualified, and who gets to decide.

Underneath that sits a second failure, and it costs more. An agency reporting only on visible metrics has no way to kill anything. Chart goes up, work was good. Every post "performed." Nothing gets retired, so nothing improves, and month nine looks like month two with new dates on it.

What Growth Actually Means on LinkedIn

Growth is a weekly loop, not a strategy document. Ours has four moves: pick one variable, change one thing, let it run long enough to read, then keep it or kill it.

What makes that loop possible is volume. Our clients publish one to three posts a week from their executives' personal profiles, and we send around 200 targeted messages a week. At LinkedIn's typical reply rate of 10 to 15 percent, 200 messages produces roughly 20 to 30 replies a week. Over a few weeks, that is enough to tell a genuinely better opening message from a worse one. Over a few days it is not, which is why we do not rewrite everything every Monday. Half the discipline here is refusing to react to a sample of eleven.

The loop runs across all three pillars at once, because that is the only way Moriah works: executive personal branding, targeted outreach, and LinkedIn Ads, all pointed at one business objective. They are not three services you buy separately. A client who publishes content and activates nothing around it gets no business. A client who sends targeted outreach with nothing published behind it gets no business either. And the loop only tells you something worth knowing when all three are live, because what you learn in one pillar changes what you do in the next. Say a post theme pulls unusually good comments from operations directors. That theme is what our targeted outreach starts referencing that month.

Channel choice also keeps measurement legible. Content published from a personal profile performs roughly 5 to 10 times better than the same content from a company page, and LinkedIn replies land at 10 to 15 percent against the 1 to 3 percent cold email typically returns. Gaps that size you can read without a statistics degree.

Key Capabilities: What a B2B Growth Marketing Agency Should Test

Which audience gets the message

The most valuable variable is also the dullest. We define a target segment precisely, then test narrower cuts of it: a job title against an adjacent one, one industry against another, decision-makers against the people who brief them. Reply rate by segment is the cleanest signal in the whole program, since the same message sent to two audiences isolates the audience.

The hook on an executive's post

The first two lines decide whether anyone stops scrolling. We test openings against each other across a month of publishing, holding the underlying argument steady so the hook is the only thing that moved. The question is not how much total reach a post collected. It is whether the right job titles are the ones reading.

The content angle behind a month of posts

Above individual posts sits the theme an executive publishes under: the specific position they take in their market. Themes get several weeks, not several days, because a position needs repetition before a market registers it. Weak ones get retired, not reworded.

The opening line of a targeted outreach message

Two hundred messages a week supports real comparison between variants. We test openings, length, and whether a message references something the executive published. Variants get compared on replies from qualified people, not on total replies, and that distinction matters once you have seen how many polite non-answers a good-looking message can generate.

The ask

The offer inside a message matters more than most of the copy around it. Asking for a call, asking for a short conversation, offering something specific from your business: these produce measurably different outcomes, and the winner varies by sector. Agencies that treat the message as creative work tend to leave this one untested. It is a question with an answer.

Ad audience and creative, when ads are running

LinkedIn Ads come in when they serve the objective, not by default. Once they are running, the audience definition and the creative get tested like everything else, against the same standard: did they move qualified conversations? Cost per click is not the test.

What We Report, and What We Leave Out

We report the sequence that leads to revenue:

• Qualified reach. Which job titles, industries and companies actually showed up in the audience for your executives' posts, not how many total impressions the account collected.
• Reply rate on targeted outreach, broken out by segment and by message variant. This is where the testing shows up as numbers.
• Qualified conversations opened. Named people who match the criteria we agreed before launch.
• Meetings booked and handed to your sales team.
• Opportunities your team accepted. The one number that settles an argument, because your side scores it and we do not.
• A record of what we tested. Every variant we ran, what it returned, whether it lived or died. That is the part of a report that tells you whether you are paying for a program that learns.

What we deliberately keep off the top of a report: impressions, follower count, and engagement rate detached from who engaged. Audience growth is real and we do report it, since that audience accumulates in your executives' names and not in a platform you rent. But it decides nothing. Neither does a "lead" who turns out to be someone who clicked once.

The Reporting Rhythm

Weekly on our side. The review rhythm with you gets agreed at the start of the engagement, and it usually settles at monthly, because one week of data rarely justifies a meeting and even more rarely supports a decision.

Every review covers the same three things: the numbers above, what we changed since last time and what it did, and what we are testing next. Then, at a longer interval, the harder conversation. Is the business objective still the right one? A program pointed at qualified leads and a program pointed at visibility to institutional investors are built differently, and objectives do occasionally move.

Who This Is For

• CEOs of established B2B companies who have paid for marketing activity before and could not tell, at the end of it, what had actually worked
• CMOs and communication directors who own a pipeline number and need a channel whose results they can defend in a board meeting
• Companies in business services, manufacturing, transport, logistics and similar sectors whose buyers are genuinely active on LinkedIn
• Leadership teams willing to publish from a personal profile. That is a real requirement on our side, not a preference
• Businesses with a settled value proposition and real trading history. Very early companies are not a fit for this model, because there is not yet a stable position to test against.

How It Works

1. Discovery, and one written objective. We go deep on what you sell and who buys it, then agree a single business objective and the definition of a qualified conversation. Nothing launches until that definition exists in writing. This is also where we confirm your buyers are actually present on LinkedIn, and we decline engagements where they are not.
2. Baseline and audience definition. We establish where you stand today on the measures above, and define the target segment precisely enough to cut it into testable slices.
3. The engine launches on all three pillars. Executive personal branding starts publishing, targeted outreach begins at volume, LinkedIn Ads come in where they serve the objective.
4. The weekly loop starts. One variable at a time, run long enough to read, kept or killed on qualified replies.
5. Monthly review, then adjust the mix. The balance between pillars shifts with the evidence. Where your buyers are not publishing much themselves, we lean harder on targeted outreach and use content to support it.

Results You Can Expect

Pipeline is not the first thing that changes. What changes first is that you know which audience, which position and which ask work for your business. That knowledge stays with you whether or not you keep us.

The second change shows up in reply rates, and it comes from sequencing, not cleverness. Targeted outreach landing on someone who has already seen your executive in their feed is not arriving cold, and LinkedIn replies already sit far above what cold email returns. Clients including ESME, Adviso Partners, Foghorn Therapeutics, Vinci Technologies and AWI run this engine with us, and our published case studies carry the specific figures per client and per sector.

The honest caveat: this needs enough time to gather real data. Anyone promising a transformed pipeline in three weeks is describing a sample too small to trust. What we can show you early is whether your audience is responding, and that is the signal that matters before volume does. We do not guarantee outcomes, and I would be careful with anyone who does.

Frequently Asked Questions

What does a B2B growth marketing agency do? It runs a repeating cycle of tests against a defined business objective, keeping whatever produces pipeline and retiring whatever does not. In our case the whole cycle runs on LinkedIn, across executive personal branding, targeted outreach and LinkedIn Ads, with results reported as qualified conversations and meetings instead of reach.

How is this different from a full-service B2B marketing agency? A full-service agency owns a wide surface: brand, website, events, several paid channels. We are deliberately narrow. LinkedIn only, which is what makes the measurement readable: when one channel carries the objective, you can attribute a change to something you actually did.

Do you work with B2B SaaS companies? Yes, provided the company is established with a settled offer and its buyers are active on LinkedIn. Software businesses tend to clear that second bar comfortably. Very early-stage companies are not a fit, whatever the sector, because a testing loop needs a stable position to test.

What metrics do you report on? Qualified reach inside your defined audience, reply rate by segment and by message variant, qualified conversations opened, meetings booked, opportunities your team accepted, and a log of every test we ran with its result. We agree the definition of "qualified" with you before launch, not afterwards.

Why not report impressions and follower growth? We do report audience growth, as context and not as the scoreboard. Impressions and followers can both climb for a full quarter without producing a single extra sales conversation. A metric earns its place at the top of a report by predicting revenue, and those two do not.

How long before the testing loop produces something useful? First readable results on audience and message variants arrive within the first several weeks, because around 200 targeted messages a week generates enough replies to compare two options. Content themes take longer, since a position needs repetition before a market registers it. Pipeline follows both.

What does it cost? In the United States, $4,000 per month. In the United Kingdom, £3,000 per month. In France, €3,000 per month. One retainer, all three pillars run together, executed in-house by our team. No per-post pricing, no separate tooling fee.

Is there a minimum term? No. No commitment, no lock-in, cancel whenever you want. The model is launch, measure, prove, and we would rather be judged on the monthly review than hold you to a contract. In practice most clients stay, because the monthly review keeps giving them a reason to.

Can we buy just the targeted outreach, or just the content? No, and the testing loop is why. Individual pillars exist as components, but the offer is the three run together against one objective. That is how LinkedIn performs, and a test in one pillar is only interpretable when the others are live.

Do you run anything outside LinkedIn? No. No paid search, no website builds or tests, nothing touching your email programs. We are a LinkedIn Certified Marketing Partner and LinkedIn is the whole of what we operate.

Get Started

If you have paid for marketing before and could not say afterwards what worked, the useful next step is a conversation about one objective and how it would be measured. Book a call and we will walk you through real business cases from your sector, confirm whether your buyers are genuinely active on LinkedIn, and show you what the first month of a testing loop would actually cover. If your audience is not there, we will tell you, and no engagement starts.