What Does a MarTech Consultant Actually Do?

A MarTech consultant designs the systems that connect your data, your marketing tools, and the team using them. The work spans audit, strategy, and implementation: deciding what to measure, choosing the platforms that measure it, integrating them with the rest of the stack, and handing the team a system they can run without you. It is not running campaigns. It is building the conditions in which campaigns become measurable.

That paragraph is the whole job. Everything below is the texture.

First, the one-line definition

A MarTech consultant lives in the connective tissue of the marketing stack. A marketing consultant works on the brand, the message, and the creative. A data consultant works on the warehouse, the modelling, and the reporting. A MarTech consultant works on the systems in between — the customer data platform, the tag manager, the marketing automation tool, the consent layer, the identity model, the activation pipes — and on the decisions that make those systems coherent instead of accidental.

Most marketing stacks are accidental. They were assembled one tool at a time, by different teams, over different quarters, against different priorities. The consultant's first job is usually to read the accident and decide which pieces to keep, which to retire, and which were never the right answer in the first place.

The four things a MarTech consultant actually does

Inside a real engagement, the work usually splits into four buckets. The same four buckets show up whether the engagement is four weeks or four months — what changes is how deep each one goes.

Audit the system underneath your marketing

The first deliverable in almost every engagement is a written read of the current state. Not a Powerpoint with logos on it. A tools inventory mapped to the team that owns each one. A data flow showing what enters the CDP, what activates from it, what is double-counted, what is missing. A consent and governance check that names the policies in plain language and identifies where the implementation stops matching them. A short list of gaps, over-investments, and blind spots, ranked by what each one is costing.

This is the part of the work where a MarTech architecture review earns its scope. It is also the part most consultants under-deliver — they survey, they don't read. A real audit ends with opinions, not observations.

Decide what to measure

Strategy is the most over-claimed and under-delivered word in the category. Inside MarTech consulting, strategy means a single concrete output: moving from "we want better attribution" to "here are the six events and three user properties that matter, here is what each one is for, here is who owns it, and here is what you will do with the answer."

A measurement framework that survives a re-org doesn't list every event a marketer might want. It lists the few the business will actually act on, the question each one answers, and the workflow it triggers. Most teams have ten times the events they need and a fraction of the answers. The consultant's job is to reverse that ratio. This is the layer where server-side tracking decisions, identity resolution, and consent design lock in for years — get this wrong and the rest of the stack inherits the debt.

Connect the platforms

This is the implementation layer. CDP integration. Server-side tag manager. Lifecycle infrastructure. Consent orchestration. Conversion APIs to Meta, Google, TikTok. Vendor coordination across three or four product teams who have never been in the same room.

The work is engineering-adjacent. The consultant writes the spec, sequences the build, sits in the standups, reviews the pull requests on the tag manager configuration, and signs off on the QA. The consultant does not replace the data engineer — the engineer ships the warehouse work. The consultant owns the system shape and makes sure the engineering effort lands inside a coherent architecture instead of a collection of point integrations. On a CDP implementation program this is where 60% of the calendar lives.

Hand the system back

The engagement that doesn't end is the engagement that didn't work. The final deliverable in every healthy MarTech consulting engagement is a runnable system the in-house team owns: documented architecture, runbooks for the things that break, a training session where the people who will use the platform daily are taught how to use it, and two weeks of on-call to catch the questions that surface only after the consultant has left the room.

A consultant who doesn't design for handover is selling permanence dressed as expertise. That isn't the work.

What a MarTech consultant does not do

The category is wide enough that buyers regularly hire consultants for adjacent work and end up disappointed. Three explicit non-goals worth naming:

• Run paid media campaigns. A MarTech consultant should be able to evaluate whether your conversion APIs are wired correctly and whether the audiences activating to Meta and Google are the ones the marketer thinks they are. A MarTech consultant should not be writing the bid strategy, picking the creative, or owning the spend.
• Write the brand brief or design the creative. Different discipline, different consultant. The MarTech consultant cares about the system; the brand consultant cares about the message; both are necessary, neither replaces the other.
• Build a custom analytics warehouse from scratch. Modelling production data into a Snowflake or BigQuery warehouse is a data engineering job. A MarTech consultant works on top of the warehouse, designs what flows in and out of it from marketing surfaces, and partners with the warehouse team — but doesn't replace them.
• Replace an in-house team. The math doesn't work. A consultant accelerates a team or seeds the conditions for one. A consultant who promises to be the team is selling a future re-platforming engagement.

A typical engagement, week by week

Specificity is a trust shortcut. Here is what a four-week strategy sprint typically looks like, the kind I run as part of the engagements I scope on melanys.me:

• Week 1 — Discovery. Interview every team that touches the stack: marketing, product, data, engineering, legal where the consent layer is involved. Map the current state. Ask the same question in three different rooms and listen for the disagreement.
• Week 2 — Diagnosis. Written audit. Twenty pages, not eighty. What works, what is broken, what costs money for nothing, what is one platform decision away from working.
• Week 3 to 4 — Decisions. A vendor recommendation, a measurement framework, a rollout sequence, or a target architecture, depending on what was scoped. Always one document with the recommendation on page one, the reasoning on page two, the alternatives considered and rejected on page three. The rest of the document is the receipts.
• Final week — Handover. Documented system, training session with the in-house team that will run it, two weeks of on-call after the formal end of the engagement.

Implementation engagements run longer — eight to sixteen weeks is typical for a CDP build — and add a build phase between Decisions and Handover. The shape of the bookends does not change.

When you should not hire a MarTech consultant

A few situations where the right answer is not a consultant:

• The problem is creative, not systemic. If the conversion rate is low because the offer is wrong, hiring a CDP consultant will not fix it. Hire a marketer.
• You don't have an in-house team to run the system after handover. A consultant builds the architecture; someone else runs it. If there is no "someone else," the system will gather dust six weeks after the engagement ends.
• The decision has already been made and you want validation. A consultant who is hired to confirm a decision will either produce honest pushback (which won't be welcome) or rubber-stamp it (which is the kind of consulting nobody is proud of). Don't hire a consultant for a decision you have already finished making.

How MarTech consulting fits next to other roles

A buyer comparing options usually has four shapes in front of them:

• MarTech consultant — depth on one specific decision or system, 4 to 16 weeks, written deliverables, ends with handover.
• MarTech agency — breadth across many platforms simultaneously, 3 to 12 months, mixed seniority, hour-burn billing.
• In-house Marketing Technologist — continuity over years, ramp time of three to six months, owns + iterates the system permanently.
• Data engineer / analytics engineer — owns the warehouse and the modelling layer; partners with the consultant rather than replacing them.

The full comparison of the first three options lives in the next piece in this cluster: MarTech consultant vs agency vs in-house hire. Briefly: the consultant wins when the decision is the bottleneck, the agency wins when the work is broad and simultaneous, and the in-house hire wins when the system will be load-bearing for the next three years.

What to bring to the first call

Discovery is faster when the buyer arrives ready. A few things worth assembling before the first conversation with any MarTech consultant:

• A list of every tool currently in the stack — including the ones that are paid for but not used, and the ones the marketing team thinks the data team owns and vice versa.
• The decision pending behind the engagement. This is the real input. Most engagement briefs describe a symptom; the consultant's first job is to find the decision underneath.
• The person on your team who will own the system after handover. If that person doesn't exist yet, the engagement is premature.

In return, a senior consultant should give you direct opinions instead of a buffet of options, written deliverables instead of slides, and a scope that ends. Anyone who is unwilling to commit to a written end state is selling a retainer dressed as a project.

What good looks like

Three signals that you've found the right MarTech consultant:

1. They ask better questions than you do. A 30-minute discovery call should reveal at least one thing about your stack that you hadn't framed that way yourself.
2. They turn down work that won't deliver. A consultant who never says no doesn't have a strong opinion of what the work is for.
3. They write things down. Decisions logged, scope captured, runbooks left behind. A consultant who can't show you a writing sample from a previous engagement (anonymized) is not a senior practitioner.

FAQ

Is a MarTech consultant the same as a marketing consultant?

No. A marketing consultant works on the brand, the message, the creative, and the channel mix. A MarTech consultant works on the systems and data underneath those decisions — the CDP, the tag manager, the lifecycle platform, the consent layer. Different discipline, different deliverables, often hired on the same project but for different problems.

Do I need a MarTech consultant or a CDP vendor?

The CDP vendor will sell you the platform and provide implementation services that are scoped to make their platform work. A MarTech consultant is independent of the vendor and is paid to make sure the platform works for your stack — including saying no to features you don't need or pushing back on the vendor's default implementation. On programs where the vendor team is doing the build, the consultant typically sits on your side of the table during sprint reviews and architecture decisions.

How long does a typical MarTech consulting engagement last?

A strategy sprint is usually four to six weeks. An implementation engagement is usually eight to sixteen weeks. A fractional advisor retainer runs for as long as the relationship is useful, typically six to twelve months. Anything significantly longer is either a full team replacement (a separate decision) or scope that has lost its shape.

Can I hire a MarTech consultant just for an audit?

Yes — and it is usually the highest-ROI engagement shape. A four-week audit produces a written read of the current state, a ranked list of gaps, and a concrete recommendation. Most teams discover the audit answers half their decisions even before the consultant gets to the implementation conversation.

If the system underneath your marketing is the question you're carrying, that is the work. The fastest way to find out if it fits is a short scoping call, or you can read more about my background and the engagements I've delivered.