OVO Talent
Home / Learn / How do I find influencers for my brand?

OVO field guide

How do I find influencers for my brand?

Finding influencers for your brand is a four-step sequence. Define the buyer and the one platform they use. Build a long list from free sources: your own customer and follower lists, competitor mentions, and YouTube's declared paid-promotion search filter. Cut that list on audience fit rather than follower count. Then verify each remaining creator's growth curve, comment substance and past disclosure hygiene before you contact anyone. Verification matters most, because follower counts are cheap to buy and real view counts are not.

OVO Creator Fit Evidence Check

Version 1.0 · Updated 2026-08-21

Turn creator selection into an evidence review across audience, category intent, production ability, delivery reliability, and disclosure history instead of a follower-count contest.

Inputs
  • Target buyer
  • Campaign objective
  • Distribution channel
  • Creator content history
  • Audience evidence
  1. Define the buyer signal

    Write the audience behavior that indicates category intent, including the questions, problems, and purchase language worth finding.

  2. Check audience reality

    Review growth shape, recent views, comment substance, audience geography, and language for patterns that conflict with the campaign market.

  3. Check category intent

    Find unpaid category history and purchase questions in the comments, then record the evidence for or against genuine audience interest.

  4. Check the content job

    Score the creator against the actual distribution need, separating audience delivery from production craft and reusable asset quality.

  5. Check operating fit

    Review delivery history, revision behavior, disclosure hygiene, competing partnerships, and availability before making the final shortlist.

OutputA shortlist with a visible reason for every inclusion and exclusion, plus the evidence that must be confirmed before outreach or contracting.

Decide what you are buying before you search for anyone

What counts as a good result changes who you should be looking for. A campaign meant to produce footage you will run as paid media needs people who can shoot clean, well-lit video, and follower count barely matters. A campaign meant to drive first-time purchases needs someone whose audience already buys in your category. Write one sentence describing the person you want to reach, then pick the single platform where that person spends time. Running two platforms at once splits the same hours across two lists, and both end up shallower than one would have been.

Where to actually look

Start with your own audience, which costs nothing to search and is the only source where you already know the person buys in your category. Export your customer list and your follower list, then check which of those accounts post publicly and have an audience of their own. These people already bought from you or already follow you, so the first message can be two lines and you do not have to guess whether they care about the category. Comment sections and tagged photos work the same way.

After that, mine the category. Search competitor brand names and your product terms on the platform you chose, then sort by recent. YouTube supports this directly. Its help page on paid product placements tells anyone who wants to "search for specific videos where the creator has declared a paid promotion" to use a link, and that link is youtube.com/results?sv=1. Adding a search term to that URL applies the same filter to your own query, so running a competitor's name through it surfaces creators who have taken paid work in your category and labelled it. What comes back depends on creators having set the declaration themselves, so read it as a lead source rather than a complete record.

TikTok has a parallel surface with a legal condition attached. Its Branded Content Policy states that "where required by law, your content may be added to the TikTok Commercial Content Library," and that "in some circumstances, TikTok may be required to ensure that Branded Content remains publicly available in the library even if the original content is deleted or altered." Coverage follows the disclosure law that applies to a given post rather than every branded post everywhere, so search the library for a named creator instead of assuming it holds a full history.

Where to lookCostGood forBlind spot
Your customer and follower listsFreeProven category interest, warmest first messageSmall pool, skews to existing fans
Competitor and category searchFreePeople already producing in your nicheSurfaces the same names everyone else contacts
YouTube declared-promotion filter (sv=1)FreeVideos where the creator set the paid-promotion declarationYouTube only, and it shows only what creators declared themselves
TikTok Commercial Content LibraryFreePast branded posts TikTok is legally obliged to keep listedTikTok only, and posts appear where disclosure law requires it, so coverage is patchy
Hashtag, audio and search-term miningFreeSmall accounts no database has indexedSlow and noisy
Discovery databases and marketplacesPaidVolume and demographic filteringModeled audience data, and rivals search the same index
An apply page on your own siteFree after setupPre-qualified inbound interestTakes months to fill

Start on the platform your buyer is already on

Platform choice follows the audience, and United States usage splits sharply by age. Pew Research Center's report on Americans' social media use, published January 31, 2024 from a survey of 5,733 U.S. adults, found that 83% of U.S. adults report ever using YouTube, 68% use Facebook, 47% use Instagram and 33% use TikTok. The same report found that 78% of adults aged 18 to 29 use Instagram against 15% of adults 65 and older, and that 62% of that younger group use TikTok against 10% of the same older group. Pew states that adults aged 30 to 49 and 50 to 64 fall somewhere in between on those platforms, so the report gives you the two ends of the age range and not the middle. A brand whose buyer is 65 or older would be opening on a platform one in ten of them uses.

PlatformShare of U.S. adults (Pew, 2024)
YouTube83% report ever using it
Facebook68%
Instagram47%
TikTok33%

Pew Research Center, Americans' Social Media Use, January 31, 2024, from 5,733 U.S. adults surveyed May 19 to September 5, 2023. Pew publishes the 18 to 29 against 65 and older split for Instagram and TikTok only, and says the middle age groups fall in between.

Cut a hundred names down to a dozen

A long list becomes a workable shortlist through four passes, in this order because each costs more time than the one before.

The checks that catch a bought audience

Audience fraud is visible without buying software. The growth curve is lumpy and continuous when it is real: a viral post produces a steep week, then growth resumes at a lower rate. A purchased batch shows up as a vertical step, tens of thousands of followers inside a day or two with nothing on the feed that accounts for it. A podcast appearance or a shoutout from a much larger account can produce the same shape, so treat the step as something that needs an explanation rather than as a verdict.

Views matter more than likes here. A like takes one tap and is the cheapest engagement to fake, while views and saves take more, so a profile whose likes look healthy while its median views sit far below its follower count is worth a closer read before you contact anyone. The arithmetic behind those ratios is covered on the engagement rate page.

Comments are worth reading as text rather than counting. Comments asking buying questions, whether a product ships to a given country or works on oily skin, indicate purchase intent. Emoji strings, one-word praise, and the same generic phrases recurring across unrelated posts indicate a templated or purchased comment pool. Ask for the audience-country breakdown from inside the app as well and compare it against where you ship, since reach in countries you do not sell in is reach you cannot convert.

SignalHow to check itHow to read it
Follower growth curveScroll the account's history for vertical stepsTens of thousands added in two days with no post behind it needs an explanation, and a purchased batch is the most common one
Views against followersCompare median views on recent non-viral posts to follower countFollowers far above what views support means the follower number is inflated
Saves and shares against likesAsk for the insights screen on three recent postsHealthy likes alongside near-zero saves is worth a second pass
Comment substanceRead fifty comments as textBuying questions signal a real audience, repeated stock phrases do not
Audience geographyRequest in-app analytics, never a third-party estimateAn audience where you do not sell is reach you cannot use

What the fake-follower rule says, and what it does not

Under 16 CFR 465.8(b), buying followers became a federal rule violation in the United States on October 21, 2024. That provision, part of the FTC's Rule on the Use of Consumer Reviews and Testimonials, makes it an unfair or deceptive act or practice for anyone to "purchase or procure fake indicators of social media influence that they knew or should have known to be fake and that materially misrepresent their influence or importance for a commercial purpose." The rule defines those indicators as metrics the public uses to assess influence, "such as followers, friends, connections, subscribers, views, plays, likes, saves, shares, reposts, and comments." The FTC announced the final rule in August 2024 and said it allows the FTC to seek civil penalties against knowing violators.

What the rule means for the brand hiring the influencer is narrower than the headline suggests. During the rulemaking a trade association argued that the provision would "hold[ ] retailers vicariously liable for the actions of independent endorsers." The FTC answered in the Federal Register: "That was not the Commission's intention." It explained that banning the distribution of fake indicators was meant to reach people supplying those indicators to buyers who would misrepresent their own influence, "not causing the dissemination of social media by users of such fake indicators, e.g., by hiring influencers who happen to have fake followers," and it added a definition at 16 CFR 465.1(g) to make that limit explicit.

Hiring someone who turns out to have bought followers is not itself a violation. The purchase provision reaches the person who bought the indicators, and the FTC narrowed the distribution provision specifically so it would not reach the brand that hired them. Your real exposure from a fake audience is commercial, because you paid for reach that does not exist. You take on a second and different risk if you republish those inflated numbers as campaign results, since the FTC's endorsement guidance says likes "from non-existent people" are "clearly deceptive" and that "both the purchaser and the seller of the fake 'likes' could face enforcement action."

Read their last few sponsored posts before you contact them

Disclosure hygiene is a screening signal, because the brand carries the monitoring duty. Score a candidate's recent sponsored posts against the FTC's published rules. Disclosures 101 for Social Media Influencers states that disclosures "are likely to be missed if they appear only on an ABOUT ME or profile page, at the end of posts or videos, or anywhere that requires a person to click MORE," tells creators not to "mix your disclosure into a group of hashtags or links," and warns against "vague or confusing terms" such as "sp," "spon," or "collab." On video the disclosure "should be in the video" itself, rather than only in the description uploaded with it.

Someone whose past paid posts fail those rules will handle yours the same way, and catching it is your job. The FTC's endorsement guides FAQ tells advertisers they "need to have reasonable programs in place to train and monitor members of their network," and that if regular monitoring is too much for you, "you should probably switch to pre-approval of posts." Asking up front whether a candidate will accept pre-approval is a fast test. Creators who work with brands regularly expect the request, so a flat refusal usually means either the person has not done paid work before or does not want the post inspected before it publishes.

Both video platforms also require a native declaration. TikTok's Branded Content Policy states that "when posting Branded Content, you must enable the commercial content disclosure toggle." YouTube defines paid product placements as "pieces of content that are created for a third party in exchange for compensation" and provides its own declaration for them. Visible paid work with no platform-level label means the creator skipped a step the platform itself requires.

What to put in the first message

Most outreach gets ignored because it asks for a decision without giving the creator anything to decide on. Include the specific post of theirs you watched and why it fit, the product, whether the offer is paid or gifted, the deliverable and rough timing, and one question that is easy to answer. Leave the brand deck out. Expect most of what you send to go unanswered, which is why a shortlist has to come from a much longer list rather than from the first few names a search returns.

When doing this yourself stops working

This method works while one person can still watch every candidate and read every post themselves. Past that, the bottleneck moves off discovery. Every creator you add brings another verification pass, contract, brief, revision round, approval and payment, and none of that gets faster the way searching does once you know where to search. Brands usually bring in outside help when that coordination load stops fitting the calendar, or when they need reach in a category they have no visibility into.

Bringing in help does not move the legal exposure. The FTC's endorsement guides FAQ states plainly that "delegating part of your promotional program to an outside company doesn't relieve you of responsibility under the FTC Act."

Frequently asked questions

How many influencers should I contact to run one campaign?

Build a long list several times larger than the group you actually want. A hundred names cutting to a shortlist of a dozen after verification, and that dozen producing three who agree to work with you, is a workable shape to plan around. Reply rates on cold creator outreach are low, some who do reply will be unavailable or a poor fit on price, and a portion will fail verification once you look closely at their audience. Skipping the long list means you contract whoever answered rather than whoever fits.

Can I just use an influencer discovery tool?

Discovery databases are genuinely useful for volume and for filtering by audience demographics, with two limits worth knowing. Audience demographics in these tools are usually modeled estimates rather than figures taken from the creator's own account, so they can disagree with a creator's in-app analytics. Every rival brand in your category also searches that same index with similar filters, which is why the same names appear on everyone's list. Use a database to build the long list, then verify on the platform itself before you contact anyone.

Is it illegal to work with an influencer who has fake followers?

Working with a creator who purchased followers is not itself a violation of the FTC's fake-indicator rule. Responding to a trade association that argued the provision would hold retailers vicariously liable for the actions of independent endorsers, the FTC stated in the Federal Register that this "was not the Commission's intention" and that it was not aimed at "hiring influencers who happen to have fake followers." What 16 CFR 465.8(b) reaches, effective October 21, 2024, is the creator's own purchase, and only where they knew or should have known the indicators were fake and the misrepresentation is material. Your exposure as the brand is commercial, since you paid for reach that does not exist, plus a separate deception risk if you republish those inflated numbers as performance claims.

What do I do if a creator will not share their analytics?

Treat a refusal as one data point rather than a disqualification, then estimate from what is public. Compare median views on the last ten non-viral posts against the follower count, since views are far harder to inflate than followers. Read the comments as text and look for buying questions. If you ship to a limited set of countries, ask for the audience-country breakdown on its own instead of the whole analytics screen, which is a smaller request most creators will agree to. Someone who has done paid work before will often send a screen recording of the insights panel rather than a still image, and a recording is harder to edit.

Should I email or send a direct message?

Send it wherever the creator tells you to send it. Many list a business email address in the profile for exactly this reason, which is a signal in itself: someone who publishes one is expecting brand mail. Direct message inboxes on accounts with large followings are filtered by the platform, so a cold message there can sit unseen indefinitely. When a direct message is the only route available, keep it short enough to read fully inside the message preview, since a wall of text gets swiped away without opening.

Do follower counts matter at all?

Follower count sets a reach ceiling and predicts little else. Two accounts with identical follower counts can differ widely in views, because platform distribution now depends more on how a specific post performs than on how many people subscribed. Judge candidates on median views across recent posts, comment quality, and whether the audience matches your buyer. Follower tiers and how they compare on cost and performance are covered on the micro versus macro influencers page.

How far back should I read a creator's posts before I contact them?

Thirty posts answers what you need and takes about fifteen minutes. Beyond counting how many of them were sponsored, check whether any of the unpaid posts touch your product category, since unpaid interest is the cheapest evidence that your product will not look out of place to that audience. Then read the three most recent sponsored posts closely for disclosure, because a creator who buried the label on someone else's campaign will bury it on yours. If nothing in thirty posts is unpaid and on-category, you are buying reach rather than fit, which is a fine trade as long as you make it deliberately.

What should I verify before signing a contract?

Ask for in-app analytics screenshots covering the three most recent posts, a list of brands in your category they have worked with along with links to those posts, what share of the last ninety days was paid content, and confirmation that they will accept pre-approval of the post before it publishes. The FTC advises advertisers that "it's much easier to review posts before they're posted than to search for them afterwards," so a refusal on pre-approval is worth weighing against everything else you found.

Sources

Planning a creator campaign? See how the OVO team scopes and delivers it.

Plan a campaign with OVO

Ready now? Start a campaign inquiry

Inquiries are reviewed personally by the founding team, and every one gets a reply within 48 hours.

Privacy choices

OVO uses limited first party analytics to understand which pages and inquiry steps are useful. Forms and the rest of the site still work when this is off.