GTM Teams: Who Owns Which Decision, and Where the Handoffs Leak

GTM Teams: Who Owns Which Decision, and Where the Handoffs Leak

Fourteen people carry go-to-market in their job title at a Series B company, and when you ask which one decides that a particular account gets worked this quarter, you get four different answers and one shrug. Nobody is underperforming. The org chart simply never assigned that decision, so it gets made repeatedly, informally, by whoever is closest to the list on any given Tuesday.

That is the characteristic failure of a go-to-market team, and it does not look like a failure from inside. Every function is busy, every dashboard has numbers on it, and the quarterly review can produce a plausible account of what happened. What is missing is a small set of named decisions with a single owner each, which is a structural property rather than a cultural one.

What a GTM team is, as distinct from sales and marketing

A go-to-market team is the group responsible for converting a market definition into revenue: marketing, sales, sales development, revenue operations, customer success, and increasingly a technical operator who builds the systems the rest of them run on. The label became popular because those functions stopped being separable. A signal picked up by ops changes the list that sales development works, which changes the message marketing writes, which changes what the account executive walks into.

Grouping them under one name is useful when it comes with shared ownership of a number. It is decorative when it means the same five functions kept the same five roadmaps and started attending one more meeting. Naming the team does not merge the incentives underneath it.

The decisions, and the one owner each needs

Strip the operating rhythm down to the things that must be decided before anyone can act, and there are five. The point of the list is not its content, which most teams would recognise. The point is that each line needs exactly one name beside it.

The decisionWhat has to be settled

• Who is in the target set this quarter
• What problem the first touch names
• Which channel carries the volume
• What makes a meeting worth taking
• What we stop doing if a number misses

Failure with no ownerWhat happens instead

• The list defaults to whoever is easiest to find data for
• Each rep improvises a different problem statement
• Every channel gets a third of the effort
• Meetings are counted, not qualified
• Nothing stops, and the next quarter inherits it

Usual ownerNot a rule, a starting point

• Revenue operations, with the ICP written down
• Product marketing
• Whoever is accountable for pipeline volume
• Sales leadership, agreed in writing with sourcing
• The GTM lead, on a stated review date

Two of those deserve emphasis because they are the ones most often left unowned.

What makes a meeting worth taking. When the people sourcing meetings and the people taking them have not agreed the criteria in writing before anything launches, the argument about quality arrives six weeks later with no way to settle it. Our own operating policy is that qualification criteria are agreed in writing before a campaign launches, and that budget, timing and authority are never treated as conditions for counting a meeting. That is a boundary worth setting whether the sourcing happens in-house or outside it.

What we stop doing. A decision with no stopping condition is a preference. The quarter cannot settle it, so it survives into the next planning cycle on the strength of whoever argues hardest.

Where the handoffs leak

The seams between functions lose more pipeline than any function loses on its own, and they lose it invisibly, because each side's dashboard shows a healthy number right up to the boundary.

1. Step 1Signal to list

   Ops finds a trigger. Nobody defines how stale a trigger may be before it is dropped, so a hiring signal from four months ago is worked as if it were live.

2. Step 2List to message

   The list carries the reason each account is on it. If that reason does not travel with the row, the message reverts to a generic pitch and the targeting work is discarded.

3. Step 3Message to meeting

   A reply arrives. Without written criteria, whoever books it decides what counts, and the definition drifts a little every week.

4. Step 4Meeting to opportunity

   The account executive learns something that would have disqualified the account at list stage. That learning goes into a call note and never reaches the person building lists.

The fourth is the expensive one. A go-to-market team without a route from closed-lost back into the targeting definition is running the same list logic in month twelve that it ran in month one, having accumulated a year of evidence about why parts of it do not work. The fix is a standing obligation on one named person to fold loss reasons into the ideal customer profile rather than a request for people to share learnings.

Alignment meetings do not close a structural gap

The standard response to leaky seams is a recurring alignment meeting. It reliably produces goodwill and unreliably produces change, for a reason that is easy to state: a meeting can surface a disagreement, but it cannot assign a decision that the org chart declined to assign. If the target-set decision belongs to nobody, a weekly forum turns it into a negotiation among four people with different incentives, which is slower than one person deciding and being wrong in public.

Three things do close the gap, and they are all boring.

A written owner per decision. One name, visible, changeable. Being wrong in an identified way is what makes a strategy correctable.

A shared artefact instead of shared meetings. One page that carries the segment, the problem, the channel, the meeting criteria and the two leading indicators, which is the same artefact a go-to-market strategy should already be producing. If the team has that page and it is current, most alignment meetings become status updates that could have been read.

A number both sides of a seam are measured on. Sourcing measured on meetings booked and sales measured on meetings that convert will argue forever. Both measured on qualified meetings that reached a second conversation will not.

The incentive seam nobody redraws

Underneath the handoffs sits a compensation structure that usually predates the team, and it quietly decides which of the five decisions actually gets made well. Marketing carried on a lead-volume target and sales carried on closed revenue are not aligned by being invited to the same standup. They are being paid to optimise two different points on the same funnel, and where those points disagree, the money wins.

The disagreement shows up in predictable places. A lead-volume target rewards a broader definition of the target set, because a narrower one costs volume immediately and pays back a quarter later. A pure closed-revenue target on the sales side rewards cherry-picking the accounts that were already going to buy, which makes the sourcing work look worse than it is. Neither behaviour is cynical. Both are the correct response to the scoreboard in front of the person.

Redrawing this does not require a compensation overhaul. It requires one shared measure sitting at the seam, with both sides carrying it: meetings that reached a second conversation, or opportunities that survived thirty days, or whatever the equivalent is in your cycle. A shared measure at the seam is what converts an argument about quality into an arithmetic problem both sides can see.

What a go-to-market team should not centralise

Consolidation has a limit, and teams that discover the coordination cost of seams sometimes overcorrect by pulling every decision to the centre. Two things belong at the edge.

The account-level judgement call. The person in the conversation knows things the list does not, and a structure that forbids them from disqualifying an account in the moment wastes the most expensive information the team collects. Centralise the definition, not the exception.

The message at the last mile. A central problem statement is what makes a hundred touches coherent. The specific sentence that names a specific company's situation cannot be written centrally at any useful scale, and attempts to do it produce copy that is technically approved and reads like nobody meant it.

The dividing line that holds up: centralise the things that must be consistent to be measurable, and push to the edge the things that must be specific to be persuasive.

The smallest team that works

Below a certain size the roles collapse into fewer people, and that is fine as long as the decisions do not collapse with them. A founder owning both the target set and the meeting criteria is a working arrangement. A target set that nobody owns because two people each assumed the other did is not.

Is every decision owned

• ~Depends: One named person can produce the current target-set definition on request
• ~Depends: The reason an account is on the list travels with the account into the message
• ~Depends: Meeting criteria exist in writing and predate the campaign
• ~Depends: Loss reasons reach the person who builds lists, on a schedule
• ~Depends: Two leading indicators have a target, a date and a stopping number
• ~Depends: Someone owns the systems layer rather than everyone maintaining their own

That last line is the one that changed most recently. Enrichment, targeting logic, sending infrastructure and routing used to be distributed across whoever was willing to learn the tool. Consolidating them under a single technical operator is now common enough to have a name and a hiring market, which the GTM engineer role covers in detail, along with the argument for why the function displaced part of the traditional sales development model.

What changes as the team grows

Adding people to a go-to-market team multiplies seams faster than it multiplies output, because every new function creates handoffs with every existing one. A team of four has six possible pairs. A team of seven has twenty-one. Those counts are simple combinatorics rather than a claim about any particular company, but they explain why the coordination cost of a go-to-market org grows faster than its headcount, and why the answer is usually fewer owned decisions rather than more forums.

The practical consequence is that growth should add capacity inside an existing seam before it adds a new one. A second person building lists is cheap to absorb. A new function with its own roadmap and its own definition of a good account is not, and it will take a quarter to discover which definition won.

The short version

A go-to-market team is a coordination structure, and coordination structures fail at their seams rather than in their boxes. Name the five decisions, put one person against each, make the target-set definition an artefact rather than a conversation, and give the seams a shared number. Everything else on the org chart is a resourcing question, which is a much easier problem than the one most teams think they have.

If the constraint you are actually hitting is volume into the top of that structure, we run the outbound half and hand back the meetings, with the qualification criteria agreed in writing before anything sends.