We find the customers who advocate for you unprompted, give them something worth posting, and measure what happens on Reddit, Instagram and Facebook against a control group that gets nothing.
| Month | Posting |
|---|---|
| Jan | 38 |
| Feb | 44 |
| Mar | 41 |
| Apr | 52 |
| May | 58 |
| Jun | 55 |
| Jul | 64 |
| Aug | 71 |
| Sep | 68 |
| Oct | 79 |
| Nov | 86 |
| Dec | 94 |
Most advocacy dies because nobody owns it. A discount code goes out, a few people use it, and nine months later nobody can say whether it worked. We treat advocacy as a channel with a roster, a content calendar and a control group.
The roster is real customers, recruited from the people already posting about you or answering questions on your behalf. They get early access, something to show, and a named person to talk to. They do not get a script.
| Month | Ambassadors |
|---|---|
| Jan | 180 |
| Feb | 265 |
| Mar | 340 |
| Apr | 470 |
| May | 520 |
| Jun | 690 |
| Jul | 810 |
| Aug | 940 |
| Sep | 1,120 |
| Oct | 1,260 |
| Nov | 1,480 |
| Dec | 1,710 |
We look for people already recommending you unprompted — reviews, comments, replies. That list is usually longer than clients expect.
Early access, something worth showing, and a direct line to a person. No scripts: a post that reads like ad copy is worth nothing on these channels.
Advocates post on their own terms. We track first posts, not signups, because a signup is not advocacy.
The posts that earn traction get supported. The rest teach us what the roster actually wants to talk about.
Reddit punishes anything that reads like marketing, which is exactly why a genuine recommendation there carries more weight than anywhere else we work. Threads keep ranking in search for years, so a good answer compounds long after it is posted.
We work through advocates who already participate in the relevant communities, always with the relationship disclosed. Our job is mostly finding the question that already has traffic and making sure the person answering it knows the subject.
| Tactic | Index |
|---|---|
| AMA threads | 92 |
| Community seeding | 74 |
| Mod partnerships | 58 |
| Niche subs | 41 |
| Tactic | Index |
|---|---|
| Story takeovers | 88 |
| Creator duets | 71 |
| Reels remix | 63 |
| Close-friends drops | 36 |
Instagram is where advocacy becomes visible to people who have never heard of you. It needs raw material — a result worth showing, a space worth filming, a before and after that is genuinely interesting.
Most of the reach comes from remixes rather than originals, so we optimise for formats other people can pick up and redo, and we measure profile visits rather than likes.
Facebook groups are the closest thing left to word of mouth at scale. You cannot buy your way in and you cannot control the room, which is the point — a recommendation there reads as a recommendation, not an ad.
We identify the groups your customers already belong to, support the advocates inside them, and co-host events that give people a concrete reason to invite someone by name.
| Tactic | Index |
|---|---|
| Group hosts | 84 |
| Event co-hosts | 66 |
| Long-form posts | 49 |
| Marketplace referrals | 33 |
Same metric, same scale, one panel per channel — so the three are read against each other rather than each on its own terms.
74k reach
48k reach
31k reach
Monthly engaged reach in thousands. All three panels share a 0–80k y-axis — that shared scale is the whole point of showing them side by side.
| Month | |||
|---|---|---|---|
| Jan | 21 | 12 | 9 |
| Feb | 26 | 15 | 11 |
| Mar | 24 | 19 | 10 |
| Apr | 33 | 17 | 14 |
| May | 38 | 24 | 16 |
| Jun | 35 | 28 | 15 |
| Jul | 47 | 26 | 19 |
| Aug | 52 | 33 | 22 |
| Sep | 49 | 37 | 20 |
| Oct | 61 | 35 | 25 |
| Nov | 66 | 42 | 27 |
| Dec | 74 | 48 | 31 |
Identifying advocates is easy and almost worthless on its own. The number that matters is the last one — people still publishing after six months.
| Stage | Ambassadors | % of identified |
|---|---|---|
| Identified | 12,400 | 100.0% |
| Activated | 3,480 | 28.1% |
| Publishing | 1,920 | 15.5% |
| Advocating | 640 | 5.2% |
Both cohorts started at the same index. The control group received no advocacy programme and no extra media — it is there so the line above it means something.
| Quarter | Ambassador cohort | Control |
|---|---|---|
| Q1 | 100 | 100 |
| Q2 | 118 | 104 |
| Q3 | 131 | 103 |
| Q4 | 152 | 109 |
| Q5 | 149 | 112 |
| Q6 | 178 | 110 |
| Q7 | 203 | 116 |
| Q8 | 236 | 121 |
If yours is not here, ask it on the call — we would rather answer it before you commit than after.
Not per post. Paying per post is what turns advocacy back into advertising, and the channels that matter here punish it. Advocates get early access, product, occasional travel and a direct line to a person. Where any compensation exists it is disclosed on the post, every time — it costs nothing and it is the difference between a programme that survives an audit and one that does not.
First posts usually land inside three weeks. A readable trend takes a quarter, because the control group needs the same amount of time before the comparison means anything. Anyone promising a curve in month one is showing you noise.
It happens, and it is usually useful. Advocates are customers, so a complaint from one is a complaint you were already getting without hearing it. We surface it through the perception indicator, respond in public where it is fair, and fix the underlying thing where it is not.
No. The roster comes from your customers, not from influencers, and the median advocate we work with has a few hundred followers. Reach comes from the number of genuine advocates and the communities they already belong to, not from any one account.
Yes, and Reddit is usually the one to start with — it is the slowest to build and the hardest to replicate, so it is worth beginning before you need it. The comparison chart above is less useful with one channel, but the funnel and the control group still hold.
A named person who can approve things inside a week, access to your review and support inboxes so we can find existing advocates, and something worth showing. The last one is the only hard requirement.
A program review is a 45-minute read of your current numbers. You leave with the three charts above, filled in with your data, whether or not we work together.