How Many Agencies Should Be on a Shortlist?

The short answer

Most formal agency reviews should reach a shortlist of three or four serious contenders, plus the incumbent only when it has a genuine chance of retaining the business. Smaller project searches may need fewer. The right number is the smallest group that gives buyers meaningful choice without sacrificing time, depth, fairness, or decision quality.

More options can feel safer. They are not always safer.

Every additional finalist creates another set of credentials to interpret, meetings to schedule, stakeholders to align, questions to normalize, references to check, and tradeoffs to remember. At some point, the process stops creating useful contrast and starts creating noise.

The objective is not to see the market. It is to make a defensible decision.

A useful default: three or four finalists

The ANA and 4As recommend beginning with a broader group for initial screening, then limiting the final shortlist to three or four agencies, with the incumbent added only when it is genuinely being considered. Their guidance also warns that a pitch does not become more effective simply because more agencies participate.

That is a strong default, but it is not a law. The correct number still depends on the assignment, the range of viable partner models, the quality of the early screening, and the decision team’s capacity to evaluate each option properly.

The operative word is not three. It is qualified.

Three poorly matched agencies do not make a good shortlist. Two highly relevant agencies may create a better decision than five generic ones. Four may be justified when the brief could credibly be solved through different partner models. Five or more finalists should require a specific reason, not stakeholder anxiety.

Long list and shortlist are different jobs

A long list helps the buyer understand the field. A shortlist helps the buyer choose.

Those stages should not carry the same number of agencies or ask agencies to do the same amount of work.

The mistake is treating an RFI, an RFP invitation, and a final pitch roster as the same stage. The RFI is a focused screening tool. The RFP asks aligned candidates to respond to the actual assignment. The final shortlist reserves the deepest evaluation for the few serious contenders.

Use two finalists when the field is already clear

A two-agency shortlist can work when:

• the assignment is tightly defined;
• the buyer has already completed meaningful screening;
• both candidates are demonstrably capable;
• the partner model is not in dispute;
• stakeholders agree on the evaluation criteria;
• the decision will turn on team, approach, working style, or commercial fit.

Two finalists create strong comparative focus. They also create less resilience if one candidate withdraws, develops a conflict, cannot meet the timeline, or reveals a material mismatch late in the process.

If the process begins with only two, the buyer should be confident that neither is there merely to validate a preferred choice.

Use three finalists for most well-defined searches

Three is often the most useful number because it creates real contrast without overwhelming the evaluation team.

It allows the buyer to compare:

• different team structures;
• different interpretations of the assignment;
• specialist versus integrated approaches;
• commercial and operating tradeoffs;
• chemistry without turning chemistry into the only criterion.

Three also makes it harder for the process to become a simple binary between the familiar option and the unfamiliar one.

If all three finalists look interchangeable, the shortlist was probably built around visibility or category labels rather than a clear definition of fit.

Use four when the brief supports genuinely different routes

Four finalists may be appropriate when the assignment can be solved through distinct, credible partner models.

For example, the buyer may need to compare:

• a specialist agency against an integrated agency;
• a boutique senior team against a larger scaled network;
• a strategy-led partner against an execution-led partner;
• a category specialist against a strong adjacent-category challenger.

The fourth agency should add a meaningful hypothesis. It should not be another version of an option already represented.

Before adding a fourth finalist, finish this sentence:

We need this agency in the process because it allows us to test whether ____.

If the team cannot complete that sentence clearly, the additional agency is probably creating volume rather than insight.

Five or more finalists need a written justification

There are legitimate exceptions: large global reviews, multiple geographic scopes, statutory procurement requirements, or searches in which several materially different capabilities are being awarded.

But when five or more agencies are being asked to complete the same final-stage process, the buyer should pause and document why.

Ask:

1. Are we still unclear about the partner model?
2. Are stakeholders adding names because the decision criteria are unresolved?
3. Are some participants included for political coverage rather than genuine consideration?
4. Are we asking the final pitch to solve work that should have happened during screening?
5. Can the decision team give every finalist equal access, attention, and feedback?

If the answer to the last question is no, the shortlist is too large for the process being run.

The five gates every finalist should pass

An agency should not reach the shortlist because it is famous, familiar, available, highly ranked, referred by someone important, or supported by an impressive reel alone.

Each finalist should pass five gates.

1. Assignment fit

The agency understands the actual problem, not only the category label. Its capabilities align with what the outside partner must own.

2. Relevant proof

The evidence connects to the assignment’s complexity, audience, operating environment, scale, or risk. A recognizable client logo is not enough.

3. Team credibility

The people expected to lead and perform the work are identifiable, available, and appropriate for the scope. The buyer is not evaluating a pitch-only cast.

4. Working-model fit

The proposed process, decision rights, handoffs, communication pattern, and specialist dependencies can work with the buyer’s internal team.

5. Commercial reality

The budget range, staffing assumptions, scope boundaries, timing, third-party costs, and likely change conditions are sufficiently clear to compare.

If an agency has not passed these gates, adding it to the final round does not increase choice. It postpones elimination.

Score before discussing preference

The strongest shortlist processes separate evidence from enthusiasm.

Before the final presentations, agree on the criteria and weighting. After each interaction, have decision-makers score independently before discussing the result as a group. Then use the conversation to understand differences in judgment, not to force mathematical certainty.

A practical scorecard might include:

Weights should change with the assignment. A highly technical implementation may place more weight on delivery, ownership, and specialist depth. A long-term strategic relationship may place more weight on senior access, operating rhythm, and chemistry.

The scorecard should make judgment more visible. It should not replace judgment.

Know when the shortlist is complete

Stop adding agencies when all of the following are true:

• every finalist could credibly win;
• each finalist represents a meaningful option;
• the decision criteria are agreed;
• the decision-makers can participate;
• the brief and budget range are clear enough for fair evaluation;
• the buyer can give each agency equivalent access and attention;
• the process can reveal team, proof, working model, and tradeoffs in depth.

If the team still does not trust the shortlist, diagnose the source of uncertainty. The answer may be a clearer brief, a different partner model, stronger screening, or better stakeholder alignment – not another agency.

The Hidden Gems perspective

Oversized searches create more work without guaranteeing better choices. The principle is simple: more is not better; the right options are better.

That is the difference between list building and partner selection.

A useful shortlist is not a miniature directory. It is a deliberately constructed set of credible choices, each there for a reason the buyer can explain.

Final shortlist checklist

Before invitations go out, confirm:

• We can explain why the search is necessary.
• We know what responsibility the partner must own.
• We have agreed on decision criteria and decision-makers.
• Every finalist has a genuine chance of winning.
• Every finalist passed the same essential screening gates.
• Each finalist adds a meaningful point of contrast.
• The incumbent is included only if retention is realistic.
• The process is proportionate to the value and scope of the opportunity.
• We can give every finalist equal access, time, and feedback.
• We are prepared to stop the process once the right partner is clear.

Frequently asked questions

Is three agencies always the right number for a pitch?

No. Three is a useful default for many well-defined searches, but two may be enough after strong screening and four may be justified when the buyer needs to compare genuinely different partner models. The correct number depends on complexity, procurement requirements, stakeholder capacity, and the quality of the candidates.

Should the incumbent be counted as an additional finalist?

Only include the incumbent when it has a realistic chance of retaining the business. If the decision to replace it has effectively been made, asking it to pitch creates unnecessary work and can distort the evaluation.

How should the RFI and RFP stages differ?

The RFI is a focused screening tool used to gather baseline information and remove obvious mismatches; it does not need a universal fixed count. ANA and 4As guidance recommends an initial RFP field of roughly 8-12 agencies, followed by a final shortlist of 3-4, plus an incumbent only when it has a realistic chance of retaining the business.

What if none of the shortlisted agencies feels right?

Do not automatically add more names. Revisit the brief, partner model, evaluation criteria, budget, and screening assumptions. A weak shortlist often indicates that the search was framed incorrectly before it indicates that the market lacks options.

Next step

If you are building or questioning an agency shortlist, start with How to Choose an Agency. If the field still feels broad, The Hidden Gems provides vetted, aligned recommendations based on the assignment, category, team, budget, timing, and risk profile.

Sources and editorial basis

1. ANA and 4As, 10 Positive Pitch Principles – shortlist size, transparency, RFI focus, team participation, and pitch process guidance.
2. ANA/4As Agency Search Simplification Report – industry pain points involving RFI standards, budget disclosure, decision-makers, chemistry, IP, and feedback.
3. IPA and ISBA, Pitch Positive Pledge – intentional, accountable pitching and reduction of unnecessary waste.
4. WFA and VoxComm, Agency Selection Guiding Principles – people-first selection, stakeholder alignment, and long-term partnership considerations.