How to judge a sub-IB's audience before you sign them
A sub-IB brings you their audience and their habits with it. Nine checks to run before you agree terms, from the follower count to what they promise the people who follow them.
The short answer
Evaluating a sub-IB means reading their audience before you agree terms, not after. Confirm every account belongs to them, record followers per account with the date, take the median post's engagement weighted across accounts, and see how concentrated the reach is. Then look for signs of bought followers, read the comments and what they promise their audience, and ask them to show the figures only they can see. Anything nobody has read is not measured, never zero.
- The follower count is where the reading starts, not what it concludes: record it per account with a date, then measure what the audience does.
- Median engagement weighted across accounts, and how concentrated the reach is, say more about a candidate than size. Signs of bought reach are questions, not verdicts.
- What a creator promises their audience travels with them into your structure. Return claims and undisclosed paid links are reasons to stop.
- Reach, countries and ages sit in the owner's analytics. Ask to see them; until you have, write not measured.
What you are judging
A sub-IB is an introducing broker you recruit into your own structure: they refer their own clients, and you earn an override on those clients' activity, paid by the broker. What you take on is their audience and their habits with it — how they grow, what they post, what they promise.
So the reading comes first. The method is the one in how to read a trading audience honestly. This guide puts it in working order, adds the checks that matter when someone will carry your link, and ends at the agreement.
Nine checks, in order
Each step ends with a check that must hold before you move on. The first eight need only public pages.
Confirm every account is theirs
Tie each account to the others by more than a name: links between them, the same website on each, or the same handle with matching content. A shared first name is not identification1. Check: every account on your list is tied to the rest.
Record followers per account, with the date
Note each account's followers, the date and where you read them, then treat the number as a ceiling, not an audience. YouTube's own help says the subscriber number is not the most accurate way to estimate an audience, because viewers subscribe to dozens of channels and may not return for every upload3. Check: every count has a date and a source; unread accounts say not measured.
Take the median post, weighted across accounts
Rank recent posts by likes plus comments, take the middle one and divide by followers: the engagement rate of a typical post. Weight each account's rate by its followers to combine them. Gravity reads up to about 30 recent posts per profile1. Check: the formula is stated, and no viral post sets the figure.
See where the reach sits
Compute audience concentration across the measured accounts. If a sub-IB's audience sits on one platform, so does your override on their clients, and one suspension removes both. Check: you know which platform carries most of the audience.
Look for bought reach
Platforms remove fake engagement they detect. Instagram began removing inauthentic likes, follows and comments from accounts using third-party apps to boost their popularity in 20184; YouTube removes subscribers gained by artificial means, such as buying them, from the total5. In the United States, the FTC's 2024 rule prohibits buying or selling fake indicators of social media influence where the buyer knew or should have known they were fake and used them for a commercial purpose6. Read each account twice, some time apart: engagement out of line with size, or a count that jumps or falls unexplained, is a question for the candidate. Check: two dated readings, and each anomaly written down as a question.
Read the comments, knowing they are curated
Comments show whether real people respond, in which language, and which broker the candidate already promotes. The creator chose what stays: YouTube lets a channel hold comments for review, block words and hide commenters7. Generic praise repeated across posts is one of the red flags. Check: you can name the recurring accounts and any current broker.
Match language, country and cadence to what you serve
Compare the language the candidate posts in, and the language comments come back in, with the languages you serve. Country is mostly invisible from outside; hold what you can infer against the countries your partner agreement allows. State a posting rate only from at least a fortnight of dated posts1. Check: each has a dated reading or says not measured.
Read what they promise their audience
A sub-IB's claims travel with them into your structure. Look for promised returns, "guaranteed" or "risk-free" wording, selected winning screenshots and undisclosed paid links: the problem claims in evidence before claims. In the UK, the FCA says influencers who promote a regulated financial product without approval from an FCA-authorised person may be committing a criminal offence8. The FTC tells influencers whose posts reach US consumers to disclose any financial relationship with a brand in the post itself9. Rules differ by country. Check: no promised returns, and paid relationships stated plainly.
Ask to see what only they can see
Reach, audience countries and age ranges sit in the owner's analytics. YouTube's Audience tab shows top geographies, age ranges and returning viewers10; Instagram's insights report reach as the unique accounts that saw a post11. Ask the candidate to show you these in their own dashboard, note the date, and hold them against the clients your broker accepts. Check: every figure is read, shown or not measured.
Two candidates, read the same way
Gravity's own worked example, illustrative: two candidates an IB might be offered in the same week.
| Candidate A | Candidate B | |
|---|---|---|
| Accounts | One Instagram account | Instagram and YouTube, linking to each other |
| Followers | 240,000 | 32,000 + 8,000 = 40,000 |
| Median interactions per post | 720 | 800 on Instagram · 320 on YouTube |
| Engagement, weighted by audience | 720 ÷ 240,000 = 0.3% | 1,120 ÷ 40,000 = 2.8% |
| Concentration (HHI) | 1.00 | 0.8² + 0.2² = 0.68 |
| Comments | Short praise, the same phrases on every post | Questions about the lesson, in the audience's language |
| Countries, ages, reach | Not measured | Not measured |
Candidate A has six times the followers and fewer interactions on a typical post: 720 against 1,120. A large following that barely responds may be bought, inactive or left over from an earlier topic: a question for step 5. Neither column yet says who would open an account.
The question no reading answers
One question stays open after every step: whether followers will open accounts and keep trading. No public reading answers it, and retention is what decides a book's worth, as how IB commission plans compare shows. Until referred clients exist, it is not measured.
From the reading to the agreement
The reading decides whether to offer terms; the partner agreement decides what they are. Your own agreement with the broker sets whether you may recruit sub-IBs, how many levels deep, and how your override is calculated. What an introducing broker is explains how that override sits beside the sub-IB's own pay. Worth having in writing before the first referral:
- The programme and plan the sub-IB joins, and how their clients are tagged to them.
- Countries and languages they may market to, within the ones your agreement allows.
- What they may say: the broker's rules on advertising and risk warnings, and no return claims.
- The next reading, and what happens if it disagrees with this one.
- The end: what happens to their clients, and to your override, if either agreement stops.
This describes common terms, not legal advice. Check the rules where you, the sub-IB and your audiences are.
Common mistakes
| Mistake | Why it misleads | Instead |
|---|---|---|
| Ranking candidates by followers | Followers are a ceiling on reach, and can be bought | Rank by median engagement, weighted across accounts |
| Using the average post | One viral post lifts it far above a typical one | Use the median post, with the formula stated |
| Treating an unread platform as zero | Zero says small; the truth is unknown | Write not measured, and ask |
| Accepting a screenshot of analytics | A picture of a number cannot be dated or checked | See the figure in the owner's dashboard |
| Skipping what they tell followers | Their claims will sit beside your programme's link | Read recent posts first; stop at return promises |
Before you sign
- Every account is tied to the candidate by more than a name.
- Followers are recorded per account, with a date and a source.
- Engagement is the median post's, weighted across accounts.
- Concentration is computed, and two dated readings exist.
- The comments are read, and any current broker is known.
- Recent posts promise no returns, and paid links are disclosed.
- Reach, countries and ages were shown in their dashboard, or say not measured.
- The terms are in writing: plan, countries, claims, the next reading and the end.
Questions
What is the difference between a master IB and a sub-IB?
How can you tell if a forex influencer has fake followers?
Should return claims rule out a sub-IB?
Sources
- How Gravity reads a profile — Gravity
- How Gravity files what it finds about a person — Gravity
- Removing Fake Likes, Follows & Comments — Instagram
- Federal Trade Commission Announces Final Rule Banning Fake Reviews and Testimonials — Federal Trade Commission
- FG24/1: Finalised guidance on financial promotions on social media — Financial Conduct Authority
- Disclosures 101 for Social Media Influencers — Federal Trade Commission
- Insights — Instagram Platform — Meta for Developers
Terms in this article
Gravity Editorial
Written by the people who build Gravity, from what the product does and what building it taught us. Figures come from Gravity's own measurements or from the source cited beside them; nothing on these pages is trading advice. See how we write.
Related in the Journal
Follower counts are the vanity figure. What an IB needs before choosing a partner is how engaged an audience is, where it sits, and what nobody has measured yet.
The referral business behind most retail forex brokers — what an IB does, how they are paid, and what makes the business last.
An IB's marketing is one system, not a feed: attention, trust, a disclosed link and a reason to stay. What each channel does, what advertising may do, and where the rules draw the line.
Brand, website and creative, made for introducing brokers.
Gravity turns an IB's own profile into a brand direction, a website and the creative they advertise with — in Arabic, Persian or English.