Case Studies vs References vs Live Work: How to Evaluate Agency Proof

Case studies, references, and live work are three different grades of agency proof. A case study shows what a firm chooses to tell you. A reference shows how it behaved when a campaign went sideways. Live work shows what it can get published and distributed today. When evaluating agency proof for retail investor marketing, weight live work first, references second, and case studies last.

Key Takeaways

• Live work is the only proof type a buyer can verify without the agency's cooperation, which is why it belongs at the top of the proof hierarchy.
• Case studies are useful for understanding process and sequencing, but they are self-selected, often anonymized, and rarely include the campaigns that underperformed.
• References become useful only when you ask behavioral questions about compliance friction, missed deadlines, and creator drop-offs instead of asking whether the client was happy.
• A paid pilot generates first-party proof faster than any reference call, and in WOLF Financial's campaign work single-month pilots commonly run $5,000 to $10,000 depending on scope and audience.

FactorCase StudiesClient ReferencesLive Work What it actually showsThe agency's preferred narrative and processHow the agency behaves over a full engagementWhat the agency can publish and distribute right now Who controls the evidenceThe agencyThe agency picks the referee, the client answersNobody, it is public Verifiable without permissionNoNoYes Selection bias riskHighModerate to highLow Main weaknessAnonymized clients, cherry-picked windows, no failuresCoached referees, short memory, no access to numbersShows output, not strategy, spend, or attribution Best use in an evaluationUnderstanding scope and workflowTesting reliability and compliance behaviorConfirming distribution capability exists today Weight in a decision20 percent30 percent50 percent

Table of Contents

• What Counts As Proof From A Retail Investor Marketing Agency?
• What Do Case Studies Actually Prove?
• What Do References Reveal That Case Studies Cannot?
• Why Is Live Work The Strongest Proof?
• The Proof Ladder: A Four-Rung Framework
• How Do You Verify Agency Proof Yourself?
• Which Proof Matters Most For Your Firm Type?
• What Are The Red Flags In Agency Proof?
• When Should You Stop Evaluating And Run A Pilot?
• Frequently Asked Questions

What Counts As Proof From A Retail Investor Marketing Agency?

Proof in an agency evaluation is evidence that a firm has already done the specific thing you are hiring it to do, for a buyer in a similar regulatory position, recently enough that the capability still exists. Three formats dominate vendor evaluation in this category: written case studies, client references, and live work you can inspect in public. They are not interchangeable, and the differences matter more in finance than in most categories because compliance review, disclosure practice, and distribution access are the real constraints.

Most buyers reverse the order. They read the deck, then call the two references the agency supplied, then sign. That sequence tests the two proof types the agency controls and skips the one it does not.

Live work: Campaign output that is publicly visible right now, such as posts, threads, Spaces recordings, video episodes, or newsletter placements you can open and read without the agency sending you anything. It matters because it is the only proof type a buyer can audit unilaterally.

One vocabulary note before the comparison. Institutional buyers write RFPs about self-directed investors, media coverage calls the same people retail investors, and regulators call them individual investors. The definition of a self-directed investor covers the same population in all three cases, so do not treat a case study about "retail campaigns" as evidence of a different capability than one about self-directed audiences.

What Do Case Studies Actually Prove?

A case study proves that an agency has a repeatable process and can describe it clearly. That is a real signal and it is not nothing. What a case study does not prove is typical outcome, because no firm publishes the engagement where the compliance team killed the creative in week three and the campaign shipped two months late.

Read case studies for structure, not for numbers. The useful information is in the mechanics: how many creators were involved, who wrote the talking points, who cleared them, how long approval took, what the reporting cadence was, and what the client did with the output afterward. That detail is hard to fabricate and it tells you whether the firm has run this workflow before.

The numbers deserve more suspicion. Impressions are the easiest metric to inflate and the least connected to net flows, ticker awareness, or holder growth. If a case study reports reach without reporting what was being measured against it, treat it as an activity log rather than a result. Buyers who want to sanity check what a defensible metric set looks like can compare against a framework for retail investor campaign metrics beyond impressions.

What case studies are good for

• Understanding scope of work and deliverable cadence
• Seeing how the firm sequences launch, education, and sustained presence
• Spotting whether compliance review is built into the workflow or bolted on
• Comparing two agencies on the same scenario during an RFP

What case studies cannot tell you

• Whether results are typical or the best of forty engagements
• What the client actually spent
• Whether the creators involved still work with the agency
• How the firm behaved when something went wrong

Anonymization is common and often legitimate. Public companies, ETF issuers, and broker-dealers frequently prohibit vendors from naming them. "A mid-size issuer" is an acceptable label. "A leading global asset manager" with no scope detail is a marketing sentence, not proof.

What Do References Reveal That Case Studies Cannot?

References reveal behavior over time: responsiveness, honesty about misses, how the agency handled legal review, and whether the senior person who pitched you stayed on the account. Those are the failure modes that actually end engagements, and none of them appear in a case study.

The catch is that the agency chooses the referee. Assume every reference you are handed is a happy client who has been briefed. That does not make the call useless, it changes the questions. Satisfaction questions produce noise. Behavioral questions produce signal.

Reference Questions That Produce Signal

• What did the agency get wrong in the first sixty days, and how did you find out about it?
• How many rounds of compliance review did creative typically take, and who fixed the language?
• Which people from the team did you actually talk to each week after month two?
• What did you ask for that they could not deliver?
• Did any creator or host drop out mid-campaign, and how was it replaced?
• What is in your reporting package that you personally use, and what do you ignore?
• If your budget were cut in half, would you keep this line item?

Ask for one reference the agency did not offer. A firm that has run dozens of programs can usually find a former client who churned for a benign reason, such as a strategy shift or a budget freeze. Willingness to make that introduction is itself a data point. Refusal is not automatically damning, since confidentiality clauses are real, but the explanation should be specific.

Why Is Live Work The Strongest Proof?

Live work is the strongest proof because you can verify it without the agency's cooperation, and because distribution capability decays. An agency that ran strong finance creator campaigns in 2023 may have lost half its roster since. Public output from the last ninety days tells you what the network can reach today, which is the only version of the network you are buying.

Live work for retail and self-directed investor campaigns is unusually easy to inspect. Creator posts, X Spaces recordings, YouTube episodes, podcast placements, newsletter sponsorships, and community AMAs are all public artifacts with timestamps. You can read the disclosure language, check whether paid placements were labeled, see the actual engagement quality in the replies, and judge whether the content was educational or promotional in tone.

What live work does not show is strategy, spend, or attribution. A visible thread tells you nothing about whether it moved platform approvals, model portfolio consideration, or holder counts. This is why live work sits at the top of the hierarchy without replacing the other two rungs: it confirms capability, while references confirm reliability and case studies explain intent. Creator-network operators like WOLF Financial can usually point to public output from the current quarter, and if a firm cannot, that gap is the finding.

One caution. Reach visible in public is not attributable performance. Any agency that connects a public post to a fund flow or a share price move without disclosing its attribution method is overclaiming, and buyers should treat that as a scoring penalty rather than a strength.

The Proof Ladder: A Four-Rung Framework

The Proof Ladder ranks agency evidence by how much of its verification depends on the agency itself. The lower the dependency, the higher the rung. Use it to score vendors consistently across an RFP instead of reacting to whoever tells the best story.

RungEvidence TypeVerification DependencyHow To Score It 1, highestLive public work from the last 90 daysNone, you can open it yourselfConfirm recency, disclosure quality, and audience fit 2References, including one you sourcedPartial, agency selects most refereesScore behavior under pressure, not satisfaction 3Documented case studies with process detailFull, agency controls the narrativeScore workflow clarity, discount the metrics 4, lowestLogos, awards, follower totals, unsourced claimsFull, and unfalsifiableTreat as context only, assign no weight

Rung four is where most vendor decks live. A wall of client logos proves someone paid an invoice once. It says nothing about scope, recency, or outcome, and in regulated finance a logo may represent a single one-off project from four years ago.

How Do You Verify Agency Proof Yourself?

Verification means confirming claims through sources the agency does not control, and it usually takes under two hours per vendor. Do it before the second call, not after the contract redline.

1. Search the agency's brand name alongside creator handles it claims to work with, and check whether recent posts exist rather than 2023 archives.
2. Open two or three named campaign artifacts and read the disclosure language. Paid promotion should carry a clear material connection disclosure consistent with FTC endorsement guidance [1].
3. Check whether any promoted security appeared with a compensation disclosure, since paid stock promotion carries disclosure obligations under Securities Act Section 17(b).
4. Look at the replies and quote posts, not the like count. Retail investor campaigns that landed produce questions about the product. Campaigns that did not produce bot replies and silence.
5. Ask for the reporting template from a real engagement with figures redacted. Template quality is a strong tell for operating maturity.
6. Ask who reviews creative before it ships and whether the firm has worked inside a broker-dealer principal approval process, which is relevant if FINRA Rule 2210 applies to your communications [2].
7. Confirm the pitch team is the delivery team, in writing, in the scope of work.

Attribution deserves its own question. Ask the agency to explain, in one paragraph, how it would connect campaign activity to an outcome you care about and where that method breaks down. Honest answers name the limits. For context on how those models are built and where they fail, an overview of attribution modeling for finance creator campaigns is a reasonable benchmark to hold vendors against.

Which Proof Matters Most For Your Firm Type?

Proof weighting shifts with what you are trying to move. An ETF issuer worried about a sub-scale fund and a pre-revenue public company defending against a short thesis need different evidence from the same vendor.

SituationProof To PrioritizeWhy It Fits ETF issuer launching or relaunching a fundLive work in the same category, plus references on compliance turnaroundTicker awareness campaigns fail on approval speed and on whether the creators can explain an index without making performance claims Public company building retail shareholder awarenessReferences from other issuers, plus live IR-adjacent workRegulation FD exposure and disclosure discipline matter more than reach, so behavior history outranks output volume Fintech platform driving funded accountsLive work plus first-party pilot dataConversion depends on audience fit that no case study can confirm, so buy a small test instead of a story Pre-launch product with no performance dataCase study process detail, then a pilotThere is nothing to attribute yet, so you are buying workflow and staged proof rather than outcomes Alternative investment manager with accredited-only audiencesReferences on compliance scopeGeneral solicitation rules constrain what can be public, so live work is thin by design and behavior history carries the decision

Some situations do not call for a creator distribution partner at all. If your problem is trade press coverage and analyst relationships, a PR firm is the better answer. If it is proxy advisory outreach, shareholder identification, or earnings logistics, an IR firm is. If you already have three internal writers and an approved content calendar, an in-house build usually beats a retainer. Comparing that decision honestly is the entire point of an agency for marketing to retail investors evaluation, and the answer is sometimes no agency.

What Are The Red Flags In Agency Proof?

The most reliable red flag is proof that cannot be checked. Everything below is a version of that problem, and each one appears regularly in finance marketing pitches.

• Percentage lifts with no baseline. "Grew engagement 340 percent" from what starting number, over what period, against what control?
• Aggregate network reach presented as campaign reach. A network with tens of millions of combined followers does not deliver tens of millions of views to your ticker.
• No named regulatory context anywhere. A firm that has run campaigns for regulated entities can name the review steps without prompting.
• Every case study is a success. Ask directly what they have stopped doing because it did not work. Silence is the answer.
• Live work older than a year. Creator rosters and platform mechanics both turn over faster than that.
• Promised outcomes. Guaranteed holder growth, guaranteed flows, or guaranteed placement counts tied to results are not deliverables a marketing partner can control.
• Undisclosed paid placements in their own portfolio. If their public work does not disclose material connections, your campaign will inherit that habit.
• Reluctance to price a pilot. Firms confident in their proof are usually willing to be tested at small scale.

A subtler one: the agency that answers proof questions with more proof of a different kind. You ask for recent live work and receive three more case studies. That substitution is the tell. Buyers running a structured process can pressure-test their scoring criteria against a broader guide to choosing a finance creator marketing agency before the shortlist call.

When Should You Stop Evaluating And Run A Pilot?

Stop evaluating once two or three vendors have cleared the live work and reference checks, because at that point additional diligence produces diminishing returns and a paid pilot produces first-party proof. A pilot converts someone else's case study into your own data, which is the only proof that fully answers your question.

Structure it so it can fail cleanly. Define one audience, one message, a fixed number of placements, a reporting format agreed in advance, and a success metric you set rather than the agency. Reasonable pilot metrics are process and engagement quality signals, not flows: did creative clear compliance inside the promised window, did the placements ship on schedule, did the audience ask product-level questions, did qualified traffic behave differently from your baseline.

On budget, agency-observed ranges are useful for planning. In WOLF Financial's campaign and proposal experience as of 2026, specialist finance marketing agencies commonly set minimum engagements around $10,000 per month, and single-month pilots commonly run $5,000 to $10,000. Investor relations marketing packages for public companies commonly run $25,000 to $50,000 per month depending on scope. These are agency-observed ranges rather than published market research, and pricing moves with audience, scope, and compliance requirements. Guidance on structuring a pilot before signing a retainer covers the contract mechanics in more depth, and a wider view of marketing to self-directed investors helps set expectations for what a single month can and cannot show.

One operating observation from campaign work across finance creator networks: the binding constraint on pilot results is almost never creative quality. It is approval latency. Programs where legal review runs longer than the content's relevance window produce technically compliant posts that arrive after the market has moved on. Test that clock during the pilot, because it will govern every month after.

Frequently Asked Questions

1. Are anonymized case studies a red flag?

No. Public companies, ETF issuers, and broker-dealers often prohibit vendors from naming them, so "a mid-size issuer" is normal. What matters is whether the anonymized study still contains scope, cadence, workflow, and compliance detail specific enough to be checked in a reference call.

2. How many references should I ask for?

Three is usually enough: two the agency supplies and one you source independently from public campaign output. The independent reference is the most informative because it was not briefed, and a firm that resists the introduction without a confidentiality reason has told you something.

3. What if an agency has no live work because all its clients require confidentiality?

That happens with accredited-investor and institutional mandates, where general solicitation limits keep output private. In that case shift weight to references and to a paid pilot, and ask the firm to walk you through a redacted reporting package instead of a public artifact.

4. Should I weight impressions in a case study at all?

Weight them as evidence of distribution scale, not of outcome. Impressions confirm a network exists and posts shipped, but they do not connect to net flows, holder growth, or funded accounts without an attribution method the agency should be able to explain and criticize.

5. Is a case study or a pilot better proof for a pre-launch product?

A pilot, once you have used case studies to confirm the workflow fits regulated content. Pre-launch firms have no performance history to attribute against, so buying a small, measurable test with agreed process metrics beats reading about someone else's launch.

6. How recent does live work need to be?

Prefer output from the last ninety days and treat anything older than a year as historical. Creator rosters, platform distribution mechanics, and community norms all change quickly, so old work confirms the agency once had a capability rather than that it has one now.

Conclusion

Evaluating agency proof is mostly a question of who controls the evidence. Case studies versus references versus live work is a hierarchy, not a menu: live work confirms capability today, references confirm behavior over time, and case studies explain intent and process. Score all three separately, verify the parts you can check without permission, then buy a small pilot and generate proof of your own.

Evaluating partners for this work? Request WOLF Financial case studies or talk to the team about scope and pricing for your situation.

References

1. FTC - The FTC's Endorsement Guides: What People Are Asking
2. FINRA - Rule 2210, Communications With The Public

Disclaimer: This article is for educational and informational purposes only. WOLF Financial is a digital marketing agency, not a registered investment adviser, broker-dealer, law firm, or compliance consultant. This content does not constitute investment, legal, tax, or compliance advice. Financial firms should consult qualified legal and compliance professionals before implementing marketing strategies.

By: Troy Lendman, WOLF Financial | About WOLF Financial