What an influencer campaign actually involves
The visible part of influencer marketing is one post. The work is everything around it, and the size of that gap is why brands who try this once in-house often do not try it twice.
End to end, a campaign runs through nine steps: a brief that says what the content must achieve and what it must not claim; a shortlist built from audience data rather than vibes; outreach; negotiation; a contract; content approval; the posting window; amplification if media budget is going behind it; and measurement. Skip one and the failure surfaces later, usually as a post you paid for and cannot use.
The admin is the real cost
To get eight creators live you generally need to approach thirty to fifty. Many never reply, and others quote above budget, are already booked with a competitor, or turn out to have an audience that is not in Singapore. Each creator who signs then generates a thread of their own: brief questions, product delivery, a shoot date, a draft, revision notes, an invoice, a posting reminder, and a chase for the live link because it went up without the tag. At twenty creators that is a part-time job for a month.
That is the honest reason influencer agencies exist in this market. It is not access, since anyone can email a creator. It is throughput, and the willingness to still be chasing politely on day nine.
The two steps that carry real risk
Disclosure is the first. Singapore advertising guidelines, administered by the Advertising Standards Authority of Singapore, require paid content to be identifiable as advertising, so put the platform label (paid partnership on Instagram, branded content on TikTok) in the contract rather than hoping the creator remembers. Claims are the second. In regulated categories such as health, aesthetics, supplements, medical services and financial products, responsibility for what a creator says in a paid post tends to land on the brand that paid for it, which is why the script gets approved before the shoot rather than after it.
One structural note: creator content is often the strongest raw material a brand gets all quarter, and most of it dies on someone else's profile inside seventy-two hours. We plan campaigns so those assets carry into your own channels and your ads, on the same calendar our social media management work runs on.
What influencer marketing costs in Singapore
Two separate numbers make up the budget, and conflating them is how brands end up surprised: what creators charge, and what it costs to have someone run the campaign. An agency quoting one blended figure is usually obscuring which is which.
What creators charge
The bands below are planning ranges, not our rate card. Creators here typically quote inside them for one deliverable with standard organic usage, but the spread within a tier is wide and a specific creator can sit outside their band for good reasons.
| Tier (followers) | Typical quote per deliverable | What the tier is good for |
|---|---|---|
| Nano (1k – 10k) | S$80 – S$400, often product plus a small fee | Volume and credibility. Best used many at a time as seeding, not as one hero post. |
| Micro (10k – 50k) | S$300 – S$1,200 | The workhorse tier here. Engagement is usually still strong and the audience is more often genuinely local. |
| Mid (50k – 200k) | S$1,000 – S$4,000 | Reach with credibility intact. Expect a media kit, a rate card and a slower reply. |
| Macro (200k – 1M) | S$3,500 – S$12,000 | Awareness at scale. Usually managed by an agent, and priced accordingly. |
| Celebrity or top-tier KOL (1M+) | S$10,000 – S$50,000 and up | Anchoring a campaign and buying PR value. Rarely efficient on a pure performance basis. |
What moves a quote, in rough order of impact: the deliverable itself, since a static grid post is not a scripted 45-second TikTok with two revision rounds; usage rights, covered below, which can double a fee on their own; exclusivity; the niche, since finance, property, technology and parenting creators quote materially higher than general lifestyle at the same follower count; and turnaround. Quoted rates are also opening positions more often than fixed prices, and three deliverables booked together almost always land below three times the single-post rate.
What agencies charge to run it
| Fee model | Typical range | What it includes, and what to watch |
|---|---|---|
| Flat monthly retainer | S$2,000 – S$6,000 a month | An ongoing programme: sourcing, screening, negotiation, contracting, approvals, posting management and reporting. Predictable, and no incentive to inflate creator spend. Check it is tied to a written number of activations a month. |
| Percentage of creator spend | Typically 15% – 25% | The same scope priced as a share of the budget. Simple and it scales, but it rewards spending more on fewer, bigger names. Ask what the fee does if you halve creator spend and double the creator count. |
| Per-campaign project fee | S$2,500 – S$8,000 per campaign | A defined launch or seasonal push with a fixed creator count. Sensible for two or three campaigns a year rather than a continuous programme. |
| No visible fee, margin on creator rates | Looks like zero | The agency marks up what the creator charges and keeps the difference. Not automatically dishonest, but you cannot judge value if you never see a creator invoice. |
The one question that changes a quote
Why follower count is the wrong primary filter
Follower count leads every media kit because it is the number that flatters. It says very little about whether a creator can move your product. Four other checks matter more, and all four cost nothing but time.
Check one: engagement rate, calculated the same way every time
Engagement rate is engagements divided by audience size, and the trap is that everybody computes it differently: against followers or against reach, counting saves and shares or not, over three posts or thirty. Pick one method, apply it to every creator on the shortlist, and compare like with like. We use engagements divided by followers over the last twelve non-paid posts, discarding the best and the worst.
| Tier | Healthy Instagram engagement rate | How to read it |
|---|---|---|
| Nano (1k – 10k) | Roughly 4% – 8% | Below about 2% at this size suggests the audience was not earned. |
| Micro (10k – 50k) | Roughly 2% – 5% | The sweet spot for most Singapore SME budgets. |
| Mid (50k – 200k) | Roughly 1.5% – 3% | A gentle decline here is normal, not a red flag. |
| Macro and above | Roughly 0.8% – 2% | Judge on reach and comment quality instead. The ratio stops carrying meaning. |
TikTok numbers are not comparable to these. Engagement there is usually measured against views rather than followers, which produces much higher percentages, and a TikTok account with modest followers can out-reach an Instagram account ten times its size. Compare TikTok creators only with other TikTok creators.
Then look past the ratio at what the comments actually are. Twenty comments asking where to buy something beats four hundred fire emojis. On Instagram, saves and shares matter more than likes, because saving a post sits closer to intent than tapping a heart.
Check two: where the audience actually lives
This is the check that matters most in Singapore and the one most often skipped. A Singapore-based creator can easily have the majority of their audience outside Singapore: content travels, the language is shared across the region, and one video that went further than expected can add tens of thousands of followers in Malaysia, Indonesia, the Philippines or India who will never walk into a shop in Bugis.
For a local restaurant, clinic, gym or service business, an overseas audience is not worth a discount. It is worth nothing. So before you pay anyone, ask for a screen recording of their own analytics showing audience by country and by city, taken in the last thirty days. A recording rather than a cropped screenshot, and live analytics rather than the media kit, which is a marketing document and is often a year out of date.
What counts as acceptable depends on your business rather than on a universal threshold. A Singapore food and beverage brand should want a clear majority of the audience in Singapore and be sceptical below about half. A regional e-commerce or B2B brand already shipping across Southeast Asia can be happy at 20%, provided the rest sits in markets it actually serves.
Check three: whether the followers are real
Bought followers are still common and still cheap. The signals: a follower graph with vertical steps rather than a slope; comments made up of generic emoji from accounts with no posts and no profile photo; an audience country mix that does not match the content or the language it is made in; engagement that stays flat while follower count climbs. Third-party audit tools give a useful authenticity estimate, but treat the score as a prompt to look harder, never as proof either way.
Check four: content fit, and whether the audience are buyers
Both answerable by scrolling. Does the creator already make the kind of content you want, in the register you want it in? Someone who has never done a product demonstration will not suddenly be good at one, and a brief that fights a creator's natural format reads as an advertisement to their audience and performs like one. Compare their sponsored posts against their organic ones too: a creator whose paid content collapses has an audience that punishes advertising.
Then, are the people following them your buyers? A following made up mostly of other creators, or mostly of students, can produce excellent numbers and no purchasing power for a S$4,000 product. Read fifty comments and click into ten of the accounts. It takes ten minutes and almost nobody does it.
Usage rights and whitelisting, in plain terms
This is the most expensive misunderstanding in influencer marketing, and the most avoidable: you pay for content, the content performs, you go to put budget behind it, and you find you never bought the right to.
The default in a bare agreement is narrow. You are paying a creator to publish something on their own channel. Unless the contract says otherwise you may not have the right to repost it, you almost certainly may not run it as an advertisement, and the creator may be free to remove it once an agreed period ends. Everything past that baseline is bought, and each layer has a price.
| What you are buying | What it lets you do | Typical uplift on the base fee |
|---|---|---|
| Organic only (the baseline) | The post lives on the creator's channel for an agreed minimum, commonly 6 or 12 months. Nothing else. | Included |
| Owned-channel reuse | Repost on your own social accounts, website, email and in-store screens for a defined term. | Roughly +20% – 40% |
| Paid usage from your account | Run the content as an ad from your own brand handle, for a defined window (30, 60 or 90 days is normal). | Roughly +30% – 100% |
| Whitelisting or partnership ads | Run ads that appear to come from the creator's handle, via Meta partnership ad codes or TikTok Spark Ads. Needs the creator to authorise it from their own account. | Roughly +30% – 80%, plus your media budget |
| Perpetuity and all media | Unlimited term, and often print, out of home and broadcast as well. | Often +100% – 300%, and frequently unnecessary |
Whitelisting is usually the highest-return line on the invoice
Whitelisting, called partnership ads on Meta and Spark Ads on TikTok, runs your paid budget through the creator's handle rather than your brand account. The ad carries their name, their picture and the social proof already on the post, and for most brands it beats the identical creative served from the brand account, because it does not announce itself as an advertisement in the same way. It needs an authorisation code generated from the creator's own account, a two-minute task they will do during a live campaign and will not do six months after being paid. Put it in the contract.
Practical rules that save money
- Buy the window you will actually use. Most brands over-buy perpetuity out of caution and under-buy the ninety days when the ad is genuinely running.
- Ask for the raw file. A downloaded TikTok export carries a watermark and a second round of compression, and it underperforms as an ad. The unwatermarked master should be a contract line.
- Price exclusivity separately. Blocking a creator from posting a competitor is a real cost to them and a real benefit to you, but it is worth buying on one or two hero creators, not across a whole campaign.
- Get the right to edit. If you may only run the content unchanged, you cannot cut a fifteen-second version, and the fifteen-second version is usually the one that works as an ad.
The mistake we see most often
How we run a campaign, and how we measure it
The sequence
A brief that is specific about the job to be done, the claims that cannot be made, and the one thing every piece of content has to communicate. Then a shortlist delivered with the screening data attached, so you approve creators on audience location and engagement rather than on a follower number and a nice grid. Then negotiation, and one contract that already contains usage, exclusivity, disclosure and the posting window, so nothing has to be reopened later.
Approvals are where campaigns lose weeks, so the shape is fixed up front: concept or script approved before anything is shot, one round of revisions on the draft, and a hard posting window. A fourth revision round is how a schedule slips by three weeks. Brief to first live post is typically four to six weeks, two of which are outreach and negotiation and cannot be compressed much, because they depend on other people replying.
What can be measured, and what cannot
Reach is a weak key performance indicator. It leads most reports because it is a large number that always goes up, and it is also the number most easily bought and the least connected to revenue. We report it, and we do not optimise for it.
- A unique discount code per creator. The cleanest attribution available if you sell online, with the caveat that it undercounts: plenty of people who buy because of a post never use the code.
- Tracked links with a distinct parameter per creator. Useful, and also an undercount, since link-in-bio traffic is a small slice of the people a post reaches.
- A campaign landing page. Lets you compare the quality of that traffic against your other channels rather than just counting it.
- A post-purchase question asking how the customer heard about you. Crude, unscientific, and often the most honest single source a Singapore SME has.
The part most reports leave out is that the majority of influence is not clickable. Someone watches a Reel, taps nothing, and searches your brand name two weeks later. Any agency claiming full attribution on influencer marketing is either running discount-code commerce only or overstating what the data can carry.
So measure lift rather than clicks. Run creators in concentrated bursts instead of spreading them evenly through the year, then compare branded search, direct traffic and enquiry volume during and just after a burst against the quiet weeks either side. Not a clean experiment, but it points the right way and gets genuinely useful after three or four bursts. If the budget allows a month with no creator activity, that gap is worth more to your measurement than one extra post.
How to choose an influencer agency, and when to skip one
The questions worth asking are the ones an agency would rather answer in a meeting than in writing. Ask them in writing.
- Do you take a margin on creator fees, and will I see the invoices? Everything else is downstream of that answer.
- Show me a creator you rejected, and why. A good agency has a rejection story with numbers in it: audience in the wrong country, engagement inconsistent with follower growth, past paid posts that flopped against their organic ones.
- Whose contract is it, and where do the usage rights sit? They should sit with you, and you should still be able to use the content if you change agency next year.
- Be careful with the word roster. Very few agencies in Singapore hold creators under exclusive contract. Most have a spreadsheet and good relationships, which is genuinely valuable, but it is not the same thing.
- What happens when a post underperforms? Reshoots, reruns and make-goods are a conversation to have before the campaign rather than after it.
When you should do this in-house instead
If you run fewer than about four creator activations a year, the coordination is small enough for a marketing executive to carry and a retainer will cost more than it saves. If your category is gifting-friendly at nano scale, meaning food and beverage, beauty or small-ticket lifestyle where creators will post for product, the work is mostly logistics and a good operations person will do it well. And if someone on your team is already active in the local creator scene, their relationships are worth more than any agency process, ours included.
An agency earns its fee on a different shape of problem: fifteen or more creators in one campaign, more than one market, a regulated category where claims have to be controlled, paid amplification on top of the organic posts, or nobody internally with time to chase eleven people for a draft.
Where we are genuinely a poor fit