How-To Guide

How do you build a creator roster?

Size a creator roster from the job it has to do, rank creators by delivered work, keep a warm bench between campaigns, and know when to stop adding names.

11 min read

Build a roster by writing down the job it has to do, then sizing the list to that job: 3 to 8 creators for a test, 5 to 12 a month for steady user content, and 15 or more for category presence. Rank every creator by delivered work, not follower count. Keep core, bench and retired tiers with a contact rhythm for each, give every name a next brief or a return date, and stop adding when no new creator can get work inside the cycle.

  • Size follows the job: a test needs a handful, a monthly content program needs five to twelve, and awareness work needs fifteen or more.
  • Rank on delivered work: shipped on time, met the brief, disclosed correctly, and rebooked.
  • Every creator on the list holds either a next brief or a return date. A name with neither goes to the bench.
  • One record per creator with terms, rights, disclosure, payment and last contact, because the list is only as good as its record.

The short answer: a standing list, not a bigger campaign

A creator roster is the list of creators a brand keeps while no campaign is running. It is not a bigger campaign list. A campaign list is built for one brief and closes when the brief closes. A roster survives the gap between briefs, which is why it needs a size, a rank and a maintenance rhythm that a campaign never needs.

The common mistake is to treat the roster as a contact list. A contact list records who exists. A roster records who is owed what: a next brief, a delivery date, a payment, a renewal of rights, or a decision that this person is no longer right for the brand. Scoop makes the structural point plainly: the roster is the centre of the program, and the brief adapts around it. Brands that keep that sentence in view stop rebuilding the creator list for every launch.

In practice a roster is built in five moves:

  1. Write the job the roster has to do in one sentence.
  2. Size the list to that job, then test the size against the coordination load.
  3. Rank each creator by delivered work rather than follower count.
  4. Set a contact rhythm and a next brief for every tier.
  5. Stop adding names when no new creator can get work inside the cycle.

None of this needs software on day one. It needs a record per creator and one decision each month. Software becomes useful when the record stops fitting in one head, and the sources place that point somewhere between fifteen and thirty active creators.

Three goal bands for a creator roster, testing at 3 to 8 creators, monthly user content at 5 to 12 a month, and category presence at 15 to 30 or more, drawn beside the coordination load markers at 5, 15 to 20 and 30 active creators.

Goal bands and load markers on one number line. Bands from Influentials, load markers from Scoop and Infmap, read 18 November 2026. The numbers are starting points for a decision, not a rule.

Superdirector

Make it work for your brand

Find it. Remake it. Post it. Superdirector turns the viral formats in your niche into your brand's carousels, X posts and video scripts, and keeps the calendar full.

Start with the job the roster has to do

Three jobs cover most rosters. One roster exists to test whether a market, a message or a format works. Another exists to keep a steady supply of user content flowing to organic channels and paid ads. The third exists to keep the brand visible across a category for a long stretch. That job decides how many creators the list needs before any other consideration.

JobWhat the roster must produceStarting band
Test a market or a messageSeveral different takes on the same brief, enough to compare3 to 8 creators
Supply user content every monthA fresh set of posts and clips for organic and paid use5 to 12 creators a month
Stay present in a categoryRepeated sightings across many feeds in the same audience15 to 30 creators or more

Influentials ties each of those bands to a goal in its own guide, and treats the numbers as starting points for a decision instead of a rule. That is the right way to read them. Shipping ten parcels a month is not the same job as a launch that needs twenty takes on one claim.

Write the job as a sentence with a countable output. Twelve creators a month, each posting one carousel, is a job. Grow on Instagram is not, and it will leave the list without a size. The sentence also settles the budget question early, because a brand that cannot pay for twelve posts a month should size the roster at the number it can pay for.

Size the roster to the goal, then to the load

A goal gives a band, and the coordination load gives the real number. Scoop describes the ladder from the operator's side: at five creators the monthly cadence is manageable in a spreadsheet, at twenty the coordinator spends significant time each month making sure the routine steps happen, and at fifty that coordination is a full-time job for someone who should be doing something more valuable.

Infmap traces the break point inside the records themselves. Notes and spreadsheets kept by different account managers start to diverge from each other past roughly 15 to 20 active creators, and past 30 the maintenance overhead of keeping views, filters and formulas consistent becomes its own operational burden. Those two readings agree closely enough to use as a load test.

For an always-on program, Scoop reports that most start with 10 to 20 creators posting monthly and scale from there. That range sits below the divergence point and above the spreadsheet limit, which is why it appears so often. With one coordinator and no shared record, start at the bottom of the band, prove the rhythm for a month, then add creators in small groups.

Compute the load once the job sentence exists. Multiply creators by posts per month to get briefs, then count the reviews, the payment runs and the rights questions each brief creates. If the calendar already needs a second person at twenty creators, the roster is too large for the team that has to run it, whatever the goal says.

Choose who stays: delivery first, reach second

A roster is a ranking, even when nobody writes the ranking down. The honest order puts delivered work first: did the creator ship on time, did the post meet the brief, did the disclosure appear where people would see it, did the audience respond, and would you brief them again. Follower count enters after those questions, because a large audience that ignores the post does not help the next campaign.

  1. On-time delivery: the post went live on the agreed day without a reminder from you.
  2. Brief fit: the claim, the format and the call to action matched what was asked.
  3. Disclosure: the paid partnership was visible with the post, not buried in a profile page.
  4. Audience response: saves, comments and clicks that a person would call real.
  5. Rebooking: the creator said yes to a second job, and did the work again.
TierWho belongsContact rhythmWhat they get next
CoreShipped twice, disclosed cleanly and rebookedEvery two weeks, plus a brief each cycleFirst look at each new brief
BenchDelivered once, or has not been used yetA monthly note with a real updateA specific date or a sample offer
RetiredMissed twice, or no longer fits the audienceOne closing noteNothing, and the record stays

GRIN sells the same shape as software. It asks brands to segment contacts so they know who to reach out to, who they are in talks with and who is already on board, and it promises the structure to manage every creator on a roster like they are the only creator on that roster. The tier names do not matter. What matters is that each tier has a different next step, because a roster where everyone gets the same message is a mailing list.

Three roster tiers as columns: core creators contacted every two weeks with first look at each brief, bench creators contacted monthly with a date or a sample offer, and retired creators closed with one note while the record stays.

The three tiers and what each one gets next. Original diagram built from the delivery checks in this article, Scoop's cadence guidance and GRIN's relationship-stage language, read 18 November 2026.

Keep creators warm between campaigns

Silence is the default failure of a roster. Campaigns end, the inbox goes quiet, and the creator who was enthusiastic in March has forgotten the brand by September. Sprout Social puts the rule in five words: do not go silent between campaigns. Its advice is to follow up with product updates and previews of upcoming launches, which gives the monthly note something to carry besides a greeting.

What the note carries should match the tier. Core creators hear about the next brief before it opens and see the results of their own last post. Bench creators get a real update and a date, so the contact is not a check-in with no purpose. GRIN tells brands to keep a birthday, a pet's name or a key milestone to hand for every creator, which is a reminder that the relationship has a memory even when the campaign does not.

Disclosure belongs in every message that asks for a post. The FTC says a creator must disclose any financial, employment, personal or family relationship with a brand, that the disclosure should be placed with the endorsement message itself, and that vague terms such as sp, spon or collab do not work. That guide adds that a creator cannot describe experience with a product they have not tried. Put that rule in the brief, then check the live post, because the creator carries the legal duty and the brand carries the reputational one.

Payment is part of warmth. Sprout Social reports that 41% of creators worldwide cite payment delays as their number one frustration when working with brands. Upfluence names missing tax forms, incomplete invoices and compliance checks as the things that slow payouts, which is a process problem the roster record can solve before the invoice arrives. Late payment is remembered longer than the brief.

Know when to stop adding

The clearest stop rule is an entry rule. A new creator joins the roster only when a brief, a sample or a date exists for them inside the current cycle. Names that pass the audience check but have no work waiting belong on a watchlist, which is a holding area for future candidates and not the roster itself. This one rule keeps the list from growing on enthusiasm alone.

Capacity sets the ceiling. Every added creator adds a brief, a review, a payment run and a rights question, and those tasks land on the same person who is already late. Scoop argues that a smaller roster of highly aligned creators posting consistently outperforms a large roster with low engagement rates. That comparison is between a list that gets worked and a list that only gets stored.

Judge the ceiling with three signals. Briefs go out late more than once in a month. Reviews pile up past the day a creator expects an answer. Payment dates slip without anyone deciding to move them. Any two of those mean the roster is full, whatever the goal band says. Freeze additions until the rhythm recovers, then add in a group small enough to absorb.

A decision path for adding a creator: does the name pass the audience check, does a brief, sample or date exist inside the cycle, does the team still send briefs on time, and does the new name go to the roster or the watchlist.

The entry rule for a new creator. Original diagram built from the delivery checks, Scoop's capacity argument and Infmap's load thresholds, read 18 November 2026. It is a decision aid, not a forecast.

Keep one record per creator and pay on time

The roster lives or dies on its record. One row per creator, in one place, with the same fields for everyone. The fields are boring and that is the point: a record that a second person can read is a record that survives a busy month.

  1. Contact and channel, including the address that actually gets an answer.
  2. Tier, with the date of the last delivered job.
  3. Agreed deliverables, rates and payment terms for the current period.
  4. Usage rights: where the content may run and for how long.
  5. Disclosure status on the most recent post.
  6. Next brief or return date, so the record carries the obligation.
  7. Last contact and what it was about.

Software can hold those fields, and the market now sells them widely. Guideflow reports, citing Influencer Marketing Hub, that 76% of brands now use a dedicated creator management platform to organize creator relationships, up from 52% in 2023. Its own criterion for a good platform is workflow coverage from discovery through outreach, campaign execution, payments and reporting, which is a useful checklist for anyone comparing tools.

Aspire and Upfluence sell the same promise from different angles: a single view of every creator and every collaboration, plus per-creator tracking so the next ranking is based on evidence. None of those products decides who stays on your roster. That decision comes from the five delivery checks and the record that carries them.

Run one quarter before you change the size

A roster is a claim about the future, so test it for a quarter before you change its size. Hold the list steady and measure four numbers: how many creators answered the last message, how many delivered on time, how many said yes to a second job, and what one delivered brief cost in fees, product and hours. Those four answers separate a roster that works from a list that merely exists.

SignalWhat it tells youWhat to do next quarter
Most creators answer the noteThe contact rhythm fits the people on the listHold the size and keep the rhythm
Deliveries land late more than onceThe brief or the review is the bottleneckFix the brief before adding anyone
Few creators accept a second jobThe offer or the fit is wrongPrune the bench and rewrite the offer
Cost per brief keeps falling as volume risesThe record and the rhythm are workingAdd a small group and watch the same numbers

Where the roster's work goes through Instagram's API, Meta's publishing guide caps an account at 100 API-published posts in a 24-hour moving period, and a carousel at 20 items. Those ceilings shape the brief more often than the roster size does.

If the posting work itself is being handed to a tool, the plan price matters to the roster budget. Superdirector's Growth plan costs $199 a month for one AI-run operated social account, and its pricing page lists two account slots with extra accounts at $99 a month each. On that plan it publishes eligible Instagram carousels and single X text posts once posting is turned on for an account. The $29 Starter plan covers work from Claude or ChatGPT without posting. Video scripts are saved with the work and never placed on the calendar.

Keep the owner's decisions separate from the tool's output. Superdirector's own terms say the owner makes the important calls about what posts, what changes and what gets paid, and that posting starts only when it is turned on for an account. A roster does not change that split. It only makes the list of people who receive those decisions explicit.

Frequently asked questions

How do you build a creator roster?

Write down the job the roster has to do, then size the list to that job: 3 to 8 creators for a test, 5 to 12 a month for steady user content, and 15 or more for category presence. Rank each creator by delivered work instead of follower count. Sort them into core, bench and retired tiers, give every name either a next brief or a return date, and stop adding when no new creator can get work inside the cycle.

How many creators should be on a roster?

The goal sets the starting band. Influentials puts testing at 3 to 8 creators, always-on user content at 5 to 12 a month, and wide awareness at 15 to 30 or more. Scoop says most always-on programs start with 10 to 20 creators posting monthly. The load then sets the real number: account managers' spreadsheets start to disagree past roughly 15 to 20 active creators, and past 30 the upkeep of views and formulas becomes its own burden.

What is the difference between a creator roster and a campaign list?

A campaign list is built for one brief and closes with it. A roster is the standing list a brand keeps between briefs, and Scoop describes it as the centre of an always-on program while the brief adapts around it. The roster carries obligations a campaign list does not: a next brief, a payment schedule, a rights renewal and a decision about who stays.

How do you keep creators interested between campaigns?

Contact each tier on its own rhythm and send something real. Sprout Social's advice is not to go silent between campaigns, and it suggests product updates and previews of upcoming launches. Core creators hear from you every two weeks and get first look at each brief. Bench creators get a monthly note with a real update and either a specific date or a sample offer.

When should you remove a creator from your roster?

Remove a creator when they miss the brief twice, when the disclosure keeps failing after one correction, or when their audience no longer matches the brand. Move them to a retired list rather than deleting the record, because the delivery history is evidence for the next decision. A single late post is not enough to remove anyone.

Do you need creator management software to run a roster?

No at small size. A record per creator and a monthly decision will carry a roster of five. Software earns its cost when the record stops fitting in one head, which the sources put somewhere between 15 and 30 active creators. Guideflow lists what such a platform covers: discovery, outreach, campaign execution, payments and reporting in one system.

How do you handle disclosures across a roster?

Put the disclosure rule in every brief and check the live post. The FTC says a creator must disclose any financial, employment, personal or family relationship with a brand, that the disclosure belongs with the endorsement message itself, and that abbreviations such as sp, spon or collab are too vague. Ownership stays with the creator, so the roster record should note who got it right.

Related practical guides

Write the job sentence before the list

Name the job, pick the band that matches it, and rank the creators you already have by what they delivered. The first month will show whether the roster is too large for the person running it.