Customer Education

Describe the Need, Not the Solution

How to explain the mission problem, outcome, constraints, and minimum requirements without accidentally locking the acquisition to a favorite product, vendor, or approach.

The fastest way to slow a package down is to hand contracting a solution before you have explained the need. The solution might be right. The vendor might be right. The brand might even be unavoidable. But if the file starts there, the CO has to work backward to figure out what problem the Government is actually trying to solve.

This lesson is about translation. Take what the customer naturally says in the hallway and turn it into language a CO can compete, justify, evaluate, and administer.

01Need vs. solution

A need describes the mission outcome the Government must achieve. A solution describes one way to achieve it.

Customers default to the solution because that's how everyday work goes. Something breaks. Someone finds a product. A vendor gives a demo. Another office says, "we use this one." Suddenly the sentence becomes "we need Vendor X" instead of "we need this capability."

Contracting has to start earlier than that. The CO needs to know the underlying problem, the outcome, the minimum performance level, and the constraints. Once those are clear, the acquisition team can decide whether the preferred solution is justified, or whether the market has other ways to meet the need.

A product name is not a requirement. A vendor name is not a requirement. A mission outcome with minimum standards is a requirement.

02The one-sentence test

Before you build the package, try to write the requirement in one sentence without naming a vendor, brand, or internal favorite. If you can't do that, the package isn't ready yet.

Use this sentence

We need [who] to [do what mission outcome] by [when], in [what environment], meeting [minimum standard], because [mission consequence].

It'll sound a little stiff. That's fine — the point isn't poetry. The point is forcing the package to name the user, outcome, timing, environment, minimum standard, and consequence.

The same elements work for supplies and construction. Swap "do" for "be delivered" or "be installed" for supplies, or "be built" or "be repaired" for construction. The user, outcome, environment, minimum, and consequence stay.

Example:

Weak: We need Acme's dashboard software.

Better: We need 85 analysts to view weekly readiness data in a web-based dashboard by Monday at 0800, using role-based access on NIPR (the Non-classified Internet Protocol Router Network), with exportable reports, because commanders use the report for Tuesday morning decisions.

The better version may still lead to Acme. But now the file explains what Acme has to do and gives the CO room to test whether anyone else can do it too.

03Outcomes, constraints, minimums, preferences

A clean requirement separates four things that often get mashed together.

Outcome

What has to happen

The mission result. Reports submitted, equipment installed, service desk staffed, system monitored, building repaired.

Constraint

What the solution must fit

Timeline, location, security, network, existing systems, physical space, law, funding, compatibility, or access limits.

Minimum

What is non-negotiable

The lowest acceptable performance level. If an offer can't meet this, it shouldn't win.

Preference

What would be nice

Useful extras, speed, polish, convenience, warranty, lower burden, better reporting, or experience beyond the minimum.

Most package fights happen because a preference is written like a minimum. "We prefer this software because the team likes it" isn't the same as "the software must ingest this existing data format without custom middleware because the Government doesn't own the interface code."

Customer move

When you catch yourself writing "must be easy," "must be best," "must be compatible," or "must be experienced," stop and define the test. Easy for whom? Best by what factor? Compatible with what version? Experienced doing what work, at what scale?

04Rewrite examples

The rewrite shows the reason underneath what you want, instead of just naming the thing you want.

Solution-first language Need-first rewrite Why it works better
We need Brand X tablets. We need 35 handheld devices that can run the inspection app offline for a full 10-hour shift, sync over Wi-Fi when back on base, and survive daily field use in rain, dust, and vehicle storage. It gives industry the performance target. If Brand X is the only realistic answer, the facts will show it.
We need Vendor Y because they know our system. We need support for an existing custom database with undocumented workflows, two legacy interfaces, and a required transition period that avoids interruption to weekly mission reporting. It explains the continuity risk without turning the vendor preference into the requirement.
We need a turnkey training package. We need a contractor to develop and deliver eight hours of role-based training for 120 users, provide student materials, record attendance, and update the content after the first delivery based on customer feedback. "Turnkey" sounds nice but hides the actual work. The rewrite names the deliverables.
We need the fastest possible delivery. We need delivery no later than 15 August because the equipment supports a scheduled inspection beginning 20 August. Late delivery requires rescheduling the inspection team and delays facility reopening. It turns urgency from emotion into a mission consequence.
We need a secure cloud tool. We need a cloud tool authorized to process CUI (Controlled Unclassified Information), available to 50 Government users on NIPR, with role-based access, audit logs, multi-factor authentication, and FedRAMP (Federal Risk and Authorization Management Program) status identified before award. "Secure" is too vague. The rewrite names the security facts the acquisition team has to handle.

05The brand-name danger zone

Brand names aren't forbidden. Sometimes a brand, model, platform, or vendor really is the only answer that fits the Government's need. The problem is when the package starts with the brand and never proves the need.

Red flags

"We have always used this." History may explain context, but it doesn't prove the current requirement.

"The team likes this one." Preference isn't a minimum need.

"It is compatible." Compatible with what exactly? Interface, form factor, firmware, warranty, data format, training burden, spare parts?

"The vendor helped us write the requirement." That can create organizational conflict of interest (OCI) issues. Tell the CO early so they can manage it.

"No one else can do it." Maybe. Bring market facts, not just confidence.

If the brand really IS the answer

Restrictions happen, and sometimes a single source genuinely is the only fit. The strongest brand-name packages document the minimum characteristics, the alternatives considered, and the evidence that supports the conclusion. Bring those facts to your CO; they'll know what to do with them and how to route the justification.

06First-draft worksheet

If you're staring at a blank page, answer these in order. Don't worry about perfect grammar yet. Contracting can work with rough facts. It can't work with silence.

1 What problem are we solving? Name the mission pain, not the preferred purchase.
2 Who uses the thing or receives the service? Users, locations, quantities, shifts, units, offices, or customer groups.
3 What has to happen when it works? Reports generated, parts delivered, calls answered, repairs completed, data transferred, students trained.
4 What are the true minimums? The capabilities, standards, delivery dates, certifications, or performance levels you can't live without.
5 What are the constraints? Security, network, data, space, schedule, access, existing equipment, warranty, licenses, funding, or legal limits.
6 What would make one acceptable offer better than another? Faster response, lower disruption, better warranty, stronger transition plan, better reporting, lower sustainment burden.
7 How will we know it's done? Acceptance criteria, deliverables, inspection owner, test method, sign-off point.

07What to send contracting

When you send the package, don't bury the real need in attachments. Give the CO a short front page that says what you're asking for and what facts are already known.

  • Need statement. One or two sentences using the formula above.
  • Known constraints. Security, schedule, compatibility, access, funding, location, or legal limits.
  • Minimum requirements. The non-negotiables, not the wish list.
  • Preferences and differentiators. What would make one acceptable approach better than another.
  • Known sources or products. Include them, but label them as market information unless you're asking for a restriction.
  • Evidence. Prior buys, quotes, product sheets, compatibility documentation, usage data, pictures, drawings, failure reports, or emails from technical owners.
Plain rule

You can tell the CO what product you think solves the problem. Just don't make the product the only description of the problem.

08What to do next

If you already have a rough requirement, run the sentence test and rewrite any vendor-first language. Then use the Requirements Builder to turn the need into package inputs. If a brand name or single source still looks necessary after the rewrite, use the justification training to stress-test the evidence.