Measuring and Optimizing Your Kingfin Referral Links in 2026

Why won't watching "clicks" alone grow your revenue?

You write an article, place your referral link, and stare at the analytics. This is where most people stop. "Clicks went up by X this month" — that feels good, but clicks are not revenue itself. A Kingfin payout is generated only when someone who clicks your referral link signs up and then trades on OlympTrade. In other words, a click is the number of people who reached the entrance — not the number who walked all the way to the goal.

Roughly, revenue looks like this: payout ≈ clicks × click-to-registration conversion (CVR) × post-registration trading. Watching only clicks means chasing just the first term of that product. If clicks double but the conversion in the middle halves, registrations don't increase. Conversely, even with the same number of clicks, lifting the conversion by 1.5× can move results closer to 1.5× without adding workload (this is only the math; actual figures vary by audience and content, and no specific CVR or amount is guaranteed).

The gap between intermediate and beginner shows up here, in "what you're watching." Beginners watch a single number — total clicks — and ride the emotional swings; intermediate affiliates break it down into "which clicks made it how far." Once you can break it down, you can see what to fix to move the needle. This article walks through that breakdown and improvement, done for real with two tools: SubID and GA4. Before the setup details, let's start by holding a map of "what to measure."

What stages can the click-to-registration funnel be broken into?

To lift conversion, you first need to capture "where people drop off" as stages. For a Kingfin referral link, the path a reader follows breaks down into roughly four stages. This becomes your map for measurement.

The key is that each stage differs in "where it can be measured" and "with which tool." Stages 1 and 2 are on your own site/social side, so they're visible in GA4 or social analytics. Stages 3 and 4 happen after the click, after moving onto Kingfin's domain, so your own GA4 can't follow them. That's the job of the Kingfin dashboard and SubID. Picture it as a dual measurement — "your site = GA4," "the result side = SubID + dashboard" — stitched together by a shared marker called SubID.

Splitting into stages clarifies where the problem lives. If "stage 2 is high but stage 3 is low," people who click are dropping off before the registration page — a sign of a problem between your article and the registration page. If "stage 1 is high but stage 2 is low," the link simply isn't being pressed — a problem with the article's content or how the button is placed. And if "stages 2 and 3 both pass but stage 4 lags," it's time to review the registered audience and the path into OlympTrade. The same "payouts aren't growing" calls for completely different moves depending on which stage stalls. That's why you break it down first.

What should you design SubID to distinguish?

SubID is a label you attach to a referral link to distinguish "which click or result it came from." For example, adding a marker like subid to the end of a link lets you view results by source on the Kingfin dashboard — "this many registrations came from this SubID." How well you design it determines the resolution of the information you'll later see.

A common beginner mistake is using the same SubID on every link. That tells you the "total result," but you'll never know "which article or placement worked." Intermediate affiliates design SubID by splitting in advance along the axes they'll want to compare later. Here are example axes worth distinguishing.

The trick to naming is to keep it simple and consistent. Decide one rule you can read back later and understand — like "article-shorthand-placement" — and stick with it throughout. Changing the rule midway makes it impossible to compare against past data, so fixing it from the start matters. For concrete SubID naming templates and a management format, the related article SubID Analysis: A Practical Template goes deeper.

Which events should you track in GA4?

If SubID is the tool for distinguishing "the source of results," GA4 is the tool for seeing "the movement on your own site up to the click." Kingfin referral links go to an external destination (kingfin.com), so the crux here is to record the outbound link click as an event. Only once you do this does funnel stage 2, "link click," become a number.

GA4 has "Enhanced measurement," with a setting that automatically captures outbound clicks. Turning this on is the first step. On top of that, planting manual events on your key CTA buttons — so a deliberate press is identifiable — raises accuracy. In fact, this very site's referral links carry an event called click_signup_button with a button_location, recording "which location's button was pressed." Bring the same idea to your own site.

The caution here is that GA4 only shows you "up to the click." After a click moves the reader onto kingfin.com, your own GA4 can't tell whether they registered (different domain). That's exactly why stages 3 onward need SubID and the dashboard — hence the division of labor. If you want to firm up GA4 setup from the basics, the related article How to Analyze an FX Affiliate Site with GA4 will help you build the foundation.

One more thing: piling on GA4 events does nothing if you haven't decided how to look at them. Deciding to monitor three things on a fixed cadence — "outbound clicks," "clicks by button," "clicks by article" — gets the operation running. For the mindset of narrowing your metrics, 3 Key Metrics for Affiliate Improvement is a useful reference.

How do you pinpoint and fix the weak stage?

Once measurement is in place, you finally read off "where it stalls." The method is simple: compute the ratio between adjacent funnel stages and look for the stage that drops sharply. The following is a model case (illustrative numbers / simulation) to show the thinking only — not real data.

"Stage 2→3 is extremely low" — clicks happen but don't reach registration — is an especially common stall. A frequent cause is that the article over-hypes and inflates expectations, so the gap with the actual registration page causes drop-off. If, at the moment of the click, the reader feels "this isn't what I thought," they won't go further. Here, aligning what the article promises with what's seen beyond it works better than tweaking button colors. Stating the facts honestly — "registration is free," "there's a Japanese dashboard" — reduces the post-click gap (hype should be avoided anyway under the Premiums Act and stealth-marketing rules).

"Stage 1→2 is low" — the link simply isn't clicked — is a placement problem. A link only at the end of the article, one buried in the body, or one without a single line on the reason to press (the benefit) — all lower the click rate. The rule for improvement is to change just one element at a time and compare before-and-after with SubID and GA4. Touch several at once and you won't know what worked. For designing the placement itself, the Kingfin Referral Link Placement Guide goes into detail — use it alongside this.

How do you connect tracking data to RevShare revenue?

The point of measurement isn't to produce a tidy report. Ultimately it's to grow your Kingfin payouts. Recall Kingfin's payout structure here. The main model is RevShare (revenue share), which compounds the longer your referrals keep trading on OlympTrade. The rate is tiered, rising with your track record, and with various bonuses combined it reaches up to 80%. Payouts are daily, withdrawable from a $10 minimum (it varies with results, and the amount is not guaranteed).

Given this structure, the direction of measurement becomes clear. RevShare changes a lot with "how long the registered person keeps trading afterward." So rather than merely increasing registrations, it pays to funnel in readers likely to keep trading, from high-quality articles. Looking at results by article with SubID, you sometimes see "an article with many registrations but little continued trading" versus "an article with few registrations but long-lasting trading." The latter is exactly the kind of asset article that pairs well with RevShare.

In other words, measurement is material for deciding "which article to write next." Once you can follow clicks, registrations, and trading, you can judge "this format works / doesn't" by data rather than gut. Pour time into the formats that work, and results compound more easily at the same workload. That's the gap between someone merely staring at clicks and someone running a measurement loop. If you want to understand the RevShare mechanism more deeply, the related articles are there too. Note that however much you optimize with measurement, results vary by individual and no specific revenue is guaranteed. Understand the mechanism and raise the odds — that's the realistic role of measurement.

What's the minimal measurement setup to build in the first two weeks?

You may feel "that's a lot to do." But you don't have to do it all at once. Narrowing the first two weeks to just creating a "measurable state" is enough. Aim for a minimal setup that runs, not a perfect analytics base.

At this size, you can build it in the gaps around your main job. What matters is not aiming for perfection from the start. Narrow your metrics, narrow your SubID axes, narrow the element you change to one. The more you narrow, the more clearly you see what worked. Just create a "measurable state" in the first two weeks, and from week 3 you simply run the same loop (monitor → hypothesize → change one element → compare). The longer you continue, the more your data becomes a personalized map for improvement.

Conversely, the biggest waste is freezing up while trying to build a perfect analytics environment from day one. The moment you get a referral link and even one click is recorded, your measurement has begun. So start from that first step — getting a referral link you can measure.

Written by

Hiro Hiraki

Editor-in-Chief, Kingfin JP. An FX affiliate specialist with over 15 years of financial and FinTech translation experience. Bilingual in Japanese and English.