
Where the buying population is a finite list of named organisations and procurement runs over years, broad marketing has nothing to work on. What fits is account-based engagement: a small number of target organisations, each researched, each with a plan, each reviewed on whether it is actually moving.
This covers how to build that list, what to know before making contact, how to say what your technology does in language the buyer can use internally, and how to tell an account that is progressing from one that is consuming effort. The constraints on what can be published are covered in the piece on defence marketing, and the internal capabilities required in the B2G operating model.
The first adaptation. A defence ministry is not an account. It contains procurement directorates, capability branches, research organisations and support agencies, each with separate budgets, separate priorities and no obligation to talk to each other.
An account is a specific buying centre or programme. Naming it that precisely changes the research, the message and the contact list, and it prevents the common situation where a company believes it has a relationship with a ministry because it knows one person in it.
Construct it from the institutional structures rather than from a purchased database.
Twenty to forty entries is realistic for a small company. A longer list is not a list, it is a way of avoiding the choice.
Most of it is public, and most companies make contact having read none of it. The difference is apparent within two minutes to the person on the other side.
Which programmes the account runs and where each sits in its lifecycle. A programme entering its final years behaves differently from one being defined.
What it has bought recently and from whom. Award notices are published. They tell you the incumbent, the value and frequently the duration.
Which frameworks it uses, because that determines whether an opportunity will be a full competition or a call-off among suppliers already admitted.
Who the incumbent is and how they are performing. Displacing a satisfactory incumbent is close to impossible. Displacing an unsatisfactory one is a different proposition, and the difference is knowable.
The account's stated priorities, from published strategy and planning documents. These are written to be read and are largely ignored by suppliers.
The most common failure in defence messaging is describing the product rather than the consequence.
A specification says what the system does. A mission statement says what the buyer gets, and it is the second that has to appear in the business case somebody writes to justify the purchase.
Product language: detection accuracy of ninety-nine point nine per cent.
Mission language: time from event to decision falls from thirty minutes to under one, so a commander acts inside the window rather than after it.
Product language: a modular architecture with open interfaces.
Mission language: the existing system stays in service five years longer, because subsystems can be replaced without requalifying the whole platform.
Product language: automated analysis of sensor data.
Mission language: the same watch is held by four people instead of twelve, and the other eight do something else.
The test is whether the sentence could appear in a document the buyer has to defend to their own finance function. Anything that could not is marketing copy, and it will be translated by somebody else or not at all.
Outside a live competition, market engagement is usually welcomed and often formally organised: supplier days, requests for information, market consultations, industry briefings. Participating in these is the single most useful activity available, because it puts you in the room while requirements are being formed.
Once a competition is live, contact is restricted, and breaching the rule can disqualify a bid. Whoever runs outreach has to know which competitions are open.
Rules on hospitality and on employing former officials also apply and vary by country. Keep engagement documented, which protects both sides and costs nothing.
One page per account. Anything longer will not be maintained.
The last item does the work. Without it a target list becomes a list of hopes that nobody is willing to abandon, and effort continues to flow to accounts that have not moved in two years.
Not activity. Meetings held and emails sent describe effort. What matters is state change.
Review the whole account set against these each quarter. The exercise usually reveals that a third of the accounts have not changed state since they were added, which is the information the review exists to produce.
A realistic pattern: a first year of understanding the organisation and becoming known, a second of participating in consultations and small instruments, a third in which something substantive becomes bidable.
Companies expecting a contract in year one generally abandon the account in year two, which is frequently the year before the work would have produced something. Setting the expectation correctly at the start is what makes the third year possible.
Usually by not competing with them.
Primes hold the integration relationships and the programme contracts, and displacement is rare. What works is occupying a capability the prime needs and does not have, being easier to work with than their internal alternative, and moving faster than a large organisation can move.
Position as a component of somebody's solution before attempting to be the solution. That is not a retreat; it is how most successful suppliers in this market started.
When the company has no capability the market wants yet. Account work then postpones a product decision, expensively.
When the addressable list is genuinely large, which occurs in dual-use categories where commercial demand dominates and the defence customer is incidental.
When the runway cannot sustain multi-year engagement. In that case a subcontract to a prime is the realistic route to revenue, and direct account development is a luxury the company cannot yet afford.
Account-based engagement fits this market because the market is a list. The work is choosing twenty to forty organisations, researching each properly using material that is already public, translating what you do into consequences the buyer can defend internally, engaging where it is permitted, and reviewing honestly whether each account has moved.
Most of that is unglamorous and none of it is fast. It is also the difference between a company that is known to the people who buy and a company that finds out about opportunities when the notice is published.

It is close to the only approach that fits. The buying population is a finite list of named organisations, procurement runs over years, and decisions are made by identifiable people in identifiable roles. That is the situation account-based methods were designed for. The adaptation required is that the account is a programme or a capability area rather than a company, since one ministry can contain a dozen unrelated buying centres with different priorities and budgets.
From the structures rather than from a database. National defence ministries and their procurement agencies, national defence innovation bodies, the European Defence Agency and the European Defence Fund for collaborative programmes, the NATO procurement and innovation organisations, and the prime contractors holding relevant programmes. Twenty to forty entries is a realistic list for a small company, and it should name the specific directorate or programme rather than the ministry.
Which programmes it runs and where they sit in their lifecycle. What it has bought recently and from whom, since award notices are published. Which framework agreements it uses. Who the incumbent supplier is and whether they are performing. What its stated priorities are in published strategy documents. Most of this is publicly available and most companies make contact without having read any of it, which is immediately apparent to the person on the other side.
By stating the operational consequence rather than the specification. Detection accuracy figures describe the product. Reduced time from event to decision, fewer people required to hold a watch, or an existing system kept in service for longer describe what the buyer gets. The test is whether the sentence could appear in a business case the buyer has to defend internally, since that document is what your claim eventually has to survive.
Contact is generally restricted once a competition is live, and breaching that can disqualify a bid. Outside a live procurement, market engagement is usually welcomed and often formally organised through supplier days and requests for information. Rules on hospitality and on employing former officials also apply and vary by country. The practical approach is to know which competitions are open before any outreach and to keep engagement documented.
Longer than commercial sales and not indefinitely. A realistic sequence is a first year of understanding the organisation and becoming known, a second year of participating in market consultations and small instruments, and a third in which a substantive opportunity becomes bidable. Companies that expect a contract in year one usually abandon the account in year two, shortly before the effort would have produced something.
By movement rather than by activity. Whether you are known to the people who matter. Whether you are invited to market consultations. Whether you have been asked for information. Whether you appear on a framework or approved supplier list. Whether the incumbent's contract end date is known to you. Each is a discrete state change, and reviewing accounts against them quarterly shows which are progressing and which have been consuming effort without moving.
Usually by not competing with them directly. Primes hold the integration relationships and the programme contracts, and a small company rarely displaces one. What works is occupying a capability the prime needs and does not have, being easier to work with than their internal alternative, and moving faster than a large organisation can. Position as a component of somebody's solution before attempting to be the solution.
The buying organisation and the specific programme. The named roles involved and what each is accountable for. The incumbent and their contract end date. The relevant framework. The planned engagement over the next four quarters. What would have to be true for a bid to be winnable. And an explicit decision date at which the account is either progressing or dropped. The last item is what stops a target list becoming a list of hopes.
When the company has no capability the market wants yet, in which case account work postpones a product decision. When the addressable list is genuinely large, which happens in dual-use categories where commercial demand dominates. And when the company lacks the runway to sustain multi-year engagement, in which case a subcontract to a prime is the more realistic route to revenue than direct account development.