Setu Cross-Border shipped 62 corridor pages from one template in 10 weeks
SEO
A worked example, not a client result. The organisation, the figures and the quotation below show how an engagement of this shape is structured and what it sets out to move. They are not outcomes achieved for a named client.
Tala SendSEO
Tala Send became the cited source in AI answers across four corridors
Content was restructured for citation rather than for ranking, then tracked weekly across four assistants until the pattern held.
4of 4 corridors cited
+64%assisted first sends
4 of 4CORRIDORS CITED
+64%ASSISTED FIRST SENDS
12 wksTO FIRST CITATION
38PAGES RESTRUCTURED
Measured over 12 weeks against a matched prior period. Attribution and confidence notes available on request.
The situation
Tala Send ranked well. Page one for their main corridors, healthy organic traffic, a content programme that had been running for two years and was genuinely good.
Then their own team started noticing something. When they asked an assistant which service was cheapest for Australia to Philippines, competitors were named and they were not, including competitors who ranked below them.
The instinct was that this was a novelty and would resolve itself. In a compare-and-decide category, being absent from the comparison is not a novelty.
What we found
Their content was written to be scanned by a human, which meant every page built to its point over several paragraphs. An answer engine extracts the first clear statement it finds, and there were none.
Definitions were embedded in clauses rather than standing as sentences. A model looking for what a corridor fee is could not find a sentence that said it.
Comparisons named no alternatives. Pages claimed to be faster and cheaper without stating faster and cheaper than what, which makes a claim unusable to a system that cross-checks.
Funnel and diagnostic table
| Stage or area | Position | Cumulative or detail | Read |
|---|---|---|---|
| Pages with an extractable definition | 0 of 38 | No liftable answer | Not citable |
| Answer-first section openings | 6 of 38 | Buried conclusions | Not extractable |
| Named comparisons | 0 | Unfalsifiable claims | Ignored by models |
| Numbers with inline scope | 2 of 19 | Bare figures | Unsupportable |
| FAQ schema present | No | Not parsed | Invisible |
What changed
1. Added a sixty-five word answer block to every page, written to stand completely alone with no reference to anything around it.
2. Restructured every section to lead with its conclusion rather than build to it.
3. Rewrote every claim to name its alternative, so the comparison became checkable rather than promotional.
4. Attached inline scope to every number, because a figure with its scope removed reads as invented to a model as well as to a compliance officer.
5. Added FAQ and Service schema, then tracked citations weekly across four assistants for the full twelve weeks.
We were explicit with Tala Send that FAQ schema would not produce a visible rich result, since those are restricted to a narrow set of site types. It stays because answer engines parse it, and that was the entire objective.
The first citation appeared in week seven on one corridor. By week twelve all four corridors were being cited by at least two of the four assistants tracked.
We were ranking above companies that assistants were recommending instead of us. Once you see that, you cannot unsee it.
— Mei Lin Tan, Head of Marketing, Tala Send
How the work ran
Answer block on every page
Sixty-five words, self-contained, tested by reading each one in isolation.
Claim rewriting with named alternatives
Every comparison made checkable rather than promotional.
Weekly citation tracking
Four assistants, four corridors, tracked for the full engagement.
What moved
| Metric | Before | After | Change |
|---|---|---|---|
| Corridors cited in AI answers | 0 of 4 | 4 of 4 | New |
| Assistants citing at least once | 0 of 4 | 4 of 4 | New |
| Pages with extractable definitions | 0 of 38 | 38 of 38 | Complete |
| Assisted first sends from organic | Baseline | +64% | Improved |
| Traditional rankings | Page 1 | Page 1 | Held |
| Numbers with inline scope | 2 of 19 | 19 of 19 | Compliant |
Mandatory line beneath: “Measured over 12 weeks against a matched prior period. Attribution and confidence notes available on request.”
The engagement in brief
Scope and shape
- Remittance app
- AUD → Philippines
- 12 weeks
- SEO
Our rankings did not move. Everything else did, which took a while to explain to the board.
— Mei Lin Tan, Head of Marketing, Tala Send
Same funnel, different corridor?
A fixed-fee growth audit covering acquisition, KYC completion, retention and tracking across your corridors. You keep the roadmap whether or not we work together.