Twelve posts a month, one page, the same destination. By month three one operator is fielding twenty real inquiries and the other is holding a few hundred likes and a tidy report. The difference is not how much was spent. It is what the money was exchanged for: a quantity of content, or a chain that can actually catch a traveller. This page does not argue about price. It breaks down where each shape of low-cost work breaks, and gives you the standard to judge any provider by — including us.
Split the birth of one inquiry into four steps and the real problem with low-cost work becomes obvious: it is rarely that the work is ugly. It is that nobody was assigned to one of the steps.
| Link | What this step solves | If it is missing | Where you see it in the numbers |
|---|---|---|---|
| ① Seen | Content reaches the recommendation feed of your source market | Only existing followers see the work — and repurposed content gets suppressed on top of that | Share of reach from non-followers, video retention, new profile visitors |
| ② Understood | It answers what travellers actually worry about: visas, payment, season, walking distance, what the price includes | A beautiful feed that never tells anyone what problem you solve | Dwell time, the specificity of the questions arriving in the inbox |
| ③ Believed | Original footage, clear pricing, verifiable credentials and references | Travellers message three or four operators; the one that reads like a template goes first | Return visits, second questions, quote requests |
| ④ Caught | A single next step, and someone answering while the traveller is awake | Links ①-③ are wasted — and this is the cheapest link to add | First response time, qualified inquiries, quote-to-booking rate |
This table also explains how "good content, no inquiries" is possible: content only carries links ① and ②. ③ and ④ live outside it. Any programme that delivers content alone can move the chain halfway at best. The mechanics of link ④ are in the WhatsApp lead handling playbook.
All three are inexpensive, but for different reasons and with different failure points. To place a provider, look at what arrives in your inbox at the end of the month.
| Shape | What it hands over monthly | Where it breaks | What you are left holding at month three |
|---|---|---|---|
| A · Content factory | A dozen finished pieces, often bulk-generated, templated or machine-translated | ① Seen and ② Understood | Posts the platform will not recommend and travellers cannot parse; page metrics possibly worse than before |
| B · Scheduler | Publishing to a frequency, replying to comments, a content calendar | ④ Caught | A page that looks maintained, and a few unanswered English messages in the inbox |
| C · Report polisher | Views, likes, follower growth in a monthly deck | Never connected to the funnel | A presentable deck, with not one number that maps to your business |
The common thread: none of the three delivers a qualified inquiry. That is not an execution failure, it is a scope failure — the thing you bought never contained that step, so no amount of effort would produce it.
Meta drew the line plainly in its March 2026 original content policy: re-uploading posts a page had no role in creating, or making low-value changes to someone else's content — adding borders, inserting captions, changing playback speed — counts as unoriginal and gets deprioritised. Enforcement escalates: reduced distribution in Feed and Reels first, then a non-recommendable status, then loss of monetisation for accounts that keep posting primarily unoriginal content.
The same announcement carries the counter-evidence: in the second half of 2025, views and watch time on original Reels on Facebook roughly doubled year on year. Distribution is being tilted toward original footage — that is Meta's own published direction, not an inference.
Search works the same way. Google's spam policy names "using generative AI tools or other similar tools to generate many pages without adding value for users" as a textbook case of scaled content abuse, and states that sites violating its policies may rank lower or not appear in results at all. Repurposing and reskinning has nowhere left to hide on either end.
Shape B's problem is not the content, it is the ending: no clear next step in the post, no page in the traveller's language, nobody answering the inbox on time. A message reading "Can you do five days in Yunnan in January for two adults?" that sits for two days is, by then, a message someone else has already quoted.
Shape C goes further: it substitutes numbers that can be achieved for outcomes that should be owned. Views can be bought, and a curiosity-driven video can produce them by accident, but neither is causally tied to someone asking your price.
Whoever's name is on the last step is who owns the result. In a contract, that name is the deliverable.
Three public benchmarks below, measured on different bases (reach rate, engagement rate, likes by page size). Do not multiply them together — read them separately. They all point the same way: in organic conditions, the audience that actually sees and reacts is an order of magnitude smaller than it looks.
| Metric | 2026 public benchmark | A page with 10,000 followers | What it proves / what it does not |
|---|---|---|---|
| Organic reach | 1.25% of followers on average; the 10-50K tier sits near 2.7% | Roughly 125-270 people per post | Proves the post was distributed; does not prove any of those people want to travel to China |
| Engagement rate | 0.15% average; 0.05% for link posts | Around 15 interactions on a follower basis | Proves the content landed; does not prove the reactors are your buyers |
| Average likes | 1-5K followers: 5-8; 10-50K: 30-36 | Around 30 likes | Proves the post is not ugly; does not prove anything commercial |
| Views | Heavily shaped by spend and recommendation systems | Anywhere from hundreds to hundreds of thousands | Proves the asset was consumed; does not prove a single contactable inquiry came of it |
So when a provider writes likes, views or follower growth into the deliverable column, it is promising something it can achieve without being accountable for your business. These are not useless numbers — they tell you whether the creative is working. They are process metrics. Process metrics belong in the middle of a report, never in the deliverable clause of a contract.
When comparing quotes you are not comparing fees. You are comparing what it costs to end up with one conversation worth having. That number can be calculated, and therefore compared.
Miss any of the three and it does not count. A bare "Hi", a visa question with no trip behind it, a WhatsApp number that does not connect — those are messages, not inquiries. Define it precisely and the cost becomes computable: cost per qualified inquiry = (fee + ad spend + costs outside the quote) ÷ qualified inquiries. Every item in the numerator should reconcile against a line in the contract, which is the practical value of pricing transparency. The boundaries of those items are laid out in the costs beyond the quote.
| Option | Quarterly outlay | Qualified inquiries | Cost per qualified inquiry | How it looks on paper |
|---|---|---|---|---|
| Content only (lowest fee) | ~USD 1,250-2,080 | 0-2 | Uncomputable / five figures | "We spent very little" |
| Content + ads, nobody catching | ~USD 3,470-4,860 | 5-10 (most lost) | ~USD 420-690 | "The ads are running" |
| Full chain (with lead handling) | ~USD 6,250-8,330 | 40-70 | ~USD 97-167 | "The most expensive one" |
The counter-intuitive row is the last one: the highest outlay carries the lowest cost per qualified inquiry, because media and lead handling are multipliers. Content is wasted the moment nobody catches it. Xingtu's measured figure on managed projects is RMB 14-32 per qualified inquiry (media included; sample is China-destination clients), which does not contradict the table above — the table uses a deliberately conservative basis: a three-month cold start, spread across total quarterly outlay. Why the two bases differ, and what the sample looks like, is set out in the pricing breakdown.
It is the denominator. When the numerator shrinks and the denominator approaches zero, the cost becomes infinite. A low fee saves on the numerator and destroys the denominator — and only the denominator turns into bookings. Ask for the denominator first, then discuss price.
Take a booking with RMB 8,000 gross margin and a one-in-eight inquiry-to-booking rate; one qualified inquiry is worth roughly RMB 1,000. A programme at RMB 1,200 per inquiry is not worth running, and one at RMB 300 deserves more budget. This beats arguing about fees, because it puts your own economics into the formula.
Price is stacked hours. Lay out what a month of this work actually involves and you can see what fits inside a fee — and what cannot.
| Monthly work item | Hours | Notes |
|---|---|---|
| Content planning and production (12 pieces) | 60-75 | At 5-6 hours each: concept, script, filming, edit, English copy, native edit |
| Ad operations and optimisation | 12-18 | Multiple test cells, creative rotation, stop-losses and scaling; review at least three times a week |
| Inbox handling and quoting | 20-30 | First response and follow-up across the source market's time zone, 1-1.5 hours a day |
| Monthly review and asset library | 5-8 | Turning winning themes, creatives and audiences into next month's input |
| Total | 97-131 | Roughly 0.6-0.8 of a full-time person, requiring content, media buying and English communication simultaneously |
At the 110-hour midpoint, a fee of RMB 3,000 a month implies RMB 27 an hour; RMB 5,000 implies RMB 45. That rate has to buy creative, paid media and English-language customer handling at once, which no labour market supports. It is achievable only by removing work items or lowering the standard of each — the first is "posting without catching leads", the second is "bulk AI copy". That is why the problem with low-cost programmes is almost always structural rather than a matter of individual effort. Hours are Xingtu's own estimate, published to explain the structure; they do not describe any other provider's pricing.
A note on how the three pricing pages divide the work: this page explains why cheap programmes fail and how to judge one; the pricing breakdown sets out what the two coaching tiers and the quarterly managed programme include, item by item; the costs beyond the quote covers what you pay outside the fee. Together they make one complete account.
After these six, it is usually clear whether a quote is solving your business or selling you activity. We answer them ourselves, and you are welcome to put them to us verbatim.
| Ask | What a vague answer sounds like | What a clear answer sounds like |
|---|---|---|
| ① Are you delivering posts, or qualified inquiries? | "We guarantee fifteen original pieces a month" | The relationship between content volume and inquiries is stated, and the definition of a qualified inquiry sits in a contract appendix |
| ② Who pays for ads, and whose account is it? | "We run the ads, it's all bundled" | Media is listed separately, spent from your own ad account, with a fee for the hours of running it |
| ③ Who answers the inbox, in which time zone, within what window? | "We reply when we see it" | A named time zone, hours, response window and holiday cover — with a real staffing plan for your source market |
| ④ Is the footage original or stock and reposts? Who owns it? | "We have plenty of material" | Sources named, commercial licensing confirmed, and copyright in the footage and edits assigned to you on exit |
| ⑤ How many numbers in the monthly report map to my business? | "Views, engagement and followers are all up" | Qualified inquiries, cost per qualified inquiry and inquiry-to-quote rate, with screenshots of lead sources |
| ⑥ If there is no qualified inquiry in three months, what do you owe me? | "Social takes time, we can't guarantee" | A quantified commitment written into the contract, with a remedy if it is missed — rather than a lecture on patience |
Question ⑥ is the dividing line. Few providers will put a quantified commitment in a contract, because it transfers risk. That is precisely what makes it the only question that separates billing for effort from billing for outcome. Our own managed-programme commitments and shortfall terms are published in the KPI section of the pricing page, for you to check line by line.
If this page only argued that cheap is always bad, it would be an advertisement. There are three situations where we actively tell clients not to buy the full chain yet.
Whatever the budget, the catching link cannot be cut. It is the cheapest part — a quote template in English, an FAQ bank, one person answering on time. It is also the most expensive to omit: everything bought before it leaks out in the last hundred metres.
So the useful question is not "can you do it cheaper". It is "at this budget, which link can wait and which one cannot". A provider willing to answer that directly is usually also willing to write the scope boundaries down.
Thirty minutes, no pitch on budget — we look for the missing link. You will come away with:
Form not loading? Open in a new window →
