Communications agencies: a practical guide for 2027

What to take away

• Define the business decision, audience change, scope, budget, and constraints before contacting firms.
• Choose relevant people and operating evidence, not the most polished credentials deck.
• Put ownership, approvals, accessibility, data rights, measurement, and exit terms in writing.
• Manage the relationship as a joint operating system with candid reviews and documented decisions.

Communications agencies provide outside skill, capacity, perspective, relationships, research, production, or geographic reach. Some lead public relations programs. Others specialize in corporate affairs, public affairs, employee communication, crisis work, media relations, content, social channels, events, research, or measurement. The label alone says little about fit. A useful choice begins with the result a business needs and the work its own team can realistically support.

Decide whether an agency solves the actual constraint

An outside firm can add scarce expertise, flexible capacity, independent challenge, or access to a market. It cannot repair unclear leadership, missing facts, a product problem, or a decision that executives refuse to make. Write the constraint in one sentence. Then test whether more staff, a specialist, a consultant, a production vendor, or a full agency is the smallest sensible answer.

Constraint	Possible outside help	Client duty that remains
No communication strategy	Senior counsel and research	Make business choices
Seasonal workload	Flexible production team	Prioritize and approve
New market	Local research and media expertise	Provide context and authority
Crisis exposure	Preparedness and surge support	Own facts and decisions
Weak measurement	Research and evaluation specialist	Name the decision evidence serves
Specialist channel	Focused creative or technical partner	Coordinate the wider program

Write the decision before the brief

State what leadership must decide after the work: enter a market, repair trust, explain a change, improve understanding, recruit a defined group, support a policy position, or prepare for an incident. Identify whose response matters and what observable change would be useful. A request for more awareness without a defined audience, baseline, or decision invites expensive interpretation.

• Business situation and why action is needed now
• Priority audiences and what is known about them
• Desired response, behavior, or decision
• Evidence already available and important unknowns
• Scope boundaries and work retained in-house
• Markets, languages, accessibility, legal, and security constraints
• Budget range, timing, decision process, and start date

Select the right engagement shape

Shape	Use it when	Watch for
Project	Outcome and end point are bounded	Unpriced revisions or rushed handoff
Retainer	Work recurs and priorities change	Vague capacity and idle months
Specialist advisory	Senior judgment matters more than output	Advice without implementation ownership
Embedded team	Daily collaboration is required	Confused management and employment boundaries
Lead agency	Several disciplines need integration	Hidden subcontracting and duplicated fees
Roster	Markets or specialties vary	Coordination cost and inconsistent standards

A large network may offer geographic coverage and deeper benches. An independent firm may provide direct senior access and fewer layers. A specialist may understand one audience or risk better than either. Size is a capacity fact, not a quality score. Ask which named people will do the work, how much time they have, and what happens when demand spikes.

Build a buying team that can make one decision

Include the budget owner, communication lead, day-to-day operator, procurement or finance partner, and any legal, privacy, security, accessibility, data, or regional specialist whose requirements could change the scope. Give one person authority to coordinate questions and one executive authority to choose. A committee that cannot agree on criteria before proposals arrive will reward presentation style instead of fit.

Role	Decision responsibility
Executive sponsor	Confirms priority and acceptable risk
Communication owner	Defines result, audience, and integration
Operator	Tests workflow, staffing, and feasibility
Procurement or finance	Tests commercial clarity and comparability
Risk specialists	Set nonnegotiable controls
Evaluation lead	Records scores, evidence, and reasons

Create a budget with a basis

Give agencies a range or a clear ceiling and say whether it includes research, paid distribution, travel, production, technology, taxes, pass-through costs, and contingency. Build an internal estimate from likely roles, hours, third-party expenses, and client effort. It is a planning reference, not proof of a fair market price. Compare proposals only after normalizing scope, seniority, assumptions, exclusions, and fee basis.

Run a fair, bounded search

Use market research to make a longlist, then screen for relevant category, audience, geography, language, risk, service, and budget experience. Remove conflicts and hard-control failures before a creative pitch. Ask a short list to answer the same core questions. Compensate substantial speculative work when appropriate, protect submitted ideas, publish the timetable, limit attendance, and tell unsuccessful firms promptly.

The PRCA's Pitch Forward best-practice principles describe a fair, realistic, ethical, structured, and healthy pitch process. The document calls for transparent objectives and budgets, appropriate shortlists, clear decision roles and criteria, respect for agency intellectual property, useful feedback, and prompt contracting. It is voluntary industry guidance, so buyers still need their own legal, procurement, and risk review.

Score evidence, people, and operating fit

Criterion	Evidence to inspect	Failure test
Problem understanding	Restated decision and assumptions	Do they challenge a faulty premise?
Relevant experience	Comparable work and named references	Is the similarity real or cosmetic?
Team	Roles, availability, location, and turnover plan	Who appears after the pitch?
Approach	Research, strategy, production, and review method	Can the method survive bad news?
Measurement	Baseline, definitions, method, and decisions	Does the plan confuse output with outcome?
Commercials	Scope, rates, expenses, caps, and change rules	Can both sides explain the invoice?
Controls	Security, privacy, accessibility, rights, and conflicts	Can claims be tested before acceptance?

Set weights before opening proposals. Score independently, discuss evidence, and record reasons. A chemistry meeting can test listening, disagreement, speed, and working style, but it should not erase missing capability. Use a paid exercise based on a bounded scenario when demonstration matters. Do not ask for a free finished campaign that the winning team may never deliver.

Check the work behind the story

Ask references about the exact office and team proposed, not only the agency brand. Confirm scope, dates, budget range, decision conditions, senior involvement, missed commitments, billing clarity, response under pressure, and handoff quality. Request a redacted plan, report, dashboard, content sample, issue log, or post-project review. Awards and testimonials are leads for diligence, not substitutes for it.

Contract the working relationship

• Deliverables, milestones, acceptance rules, and revision limits
• Named roles, minimum senior involvement, and substitution process
• Fee model, rate card, pass-through markup, expenses, taxes, and caps
• Approval authority, response times, escalation, and crisis coverage
• Conflicts, confidentiality, data access, retention, security, and incident notice
• Accessibility, substantiation, disclosures, records, and review duties
• Ownership and permitted use of drafts, research, creative work, and tools
• Measurement definitions, source access, exports, and correction process
• Termination rights, transition support, final files, and deletion confirmation

Start with a working session, not a ceremony

First 30 days	Concrete output
Days 1 to 5	Decision map, facts, access, risks, and open questions
Days 6 to 10	Audience evidence and baseline review
Days 11 to 15	Strategy choices, rejected options, and assumptions
Days 16 to 20	Work plan, approval map, measures, and controls
Days 21 to 30	Pilot output, quality review, and revised forecast

Agree on a single source of truth for the brief, schedule, approvals, decisions, risks, budget, and performance definitions. Hold a short operating meeting for work and a separate monthly review for choices, evidence, and relationship health. Bring problems early. A useful agency reports what failed, what changed, and what it recommends, while a useful client supplies facts, access, decisions, and candid feedback on time.

Measure the relationship and the result

• Outcome against a documented starting point
• Quality, accuracy, accessibility, and correction record
• Schedule and budget variance with explanations
• Decision speed and approval rework
• Audience or market learning produced
• Risks identified before they became incidents
• Team continuity and promised senior attention
• Actions taken because of the evidence

Review the relationship at 30, 90, and 180 days. At exit, transfer final assets, methods, account access, licenses, data, consent records, and unresolved risk notes.

Verify communications agencies before release

For communications agencies, the GAO evaluation design guide explains how evaluation questions, evidence needs, and design choices fit together. The guide is written for federal program evaluation. Use its design discipline as a check on the method, not as proof that a marketing result is causal or transferable.

The W3C Privacy Principles statement gives system designers a shared vocabulary for privacy and warns against shifting privacy work onto individuals. Apply that principle to the data flow behind communications agencies. It does not replace the law, contract terms, consent analysis, or a review of the actual configuration.

The GOV.UK technology selection guidance recommends choices that can change over time, preserve data control, address security risk, and include ownership cost. Those public-service rules become useful buying questions for communications agencies, but they are not private-sector mandates or product endorsements.

Apply these checks to the actual communications agencies workflow. Record the tested data, roles, product versions, exceptions, and approval date. Repeat the review after a material source, model, access, contract, or decision change. The added sources define separate evaluation, privacy, and operating questions; none certifies the local implementation or supplies a guaranteed marketing result.

Common questions

How many communications agencies should a company invite?

Use the smallest credible shortlist after market research. Three or four well-matched firms often permit comparison without imposing unnecessary pitch cost, but complexity and procurement rules may require another approach.

Should price be the deciding factor?

Price matters, but its weight should reflect how clearly the requirement can be defined and the cost of weak performance. Compare normalized scope and risk before comparing totals.

Is an agency of record always better than project firms?

No. A lead relationship helps when continuity and integration matter. Project or specialist firms can be better for bounded work, uncommon expertise, testing, or geographic needs.

Who should own agency management?

A named client lead should own priorities, access, approvals, budget visibility, and feedback. The agency should have a named counterpart with authority to manage delivery and escalation.

When should a company change agencies?

Consider change when the required capability or market changes, repeated control failures persist, promised staffing is absent, evidence is unreliable, or the relationship cannot recover after a documented correction plan.