How to Choose a Word-of-Mouth Marketing Agency: A Complete Guide to Avoiding Bad Actors
A brand marketing lead once told me his company had spent NT$180,000 on three months of word-of-mouth marketing. When he finally looked closely, he found that dozens of the “posting accounts” had exactly one post to their name — a glowing mention of his brand. No other content, no engagement history, and profile photos that were plainly AI-generated faces.
This is not an isolated case. Taiwan’s word-of-mouth marketing market is chaotic, and the barrier to entry is close to zero. A few PTT accounts and the ability to post on Dcard are enough to open shop and start taking clients. For brands, telling the good from the bad is the hardest part.
Why Fake Accounts Are Riskier Than You Think
A lot of brands figure it does not matter whether the accounts belong to real people, so long as the posts go up. That thinking has several blind spots.
PTT and Dcard both have moderation systems. (PTT is Taiwan’s largest and oldest online forum; Dcard is the community platform most popular with younger Taiwanese users.) PTT board moderators watch for accounts whose recent activity clusters around recommending a single brand, and Dcard’s algorithm down-ranks accounts with thin engagement histories. Posts published through fake accounts have a high chance of being deleted or shadowbanned within two to four weeks — which means the money you spent evaporates.
The SEO value is close to zero. A PTT thread with no replies and no discussion carries almost no weight in search. The reason PTT threads rank on the first page of Google is the page activity generated by replies and comments — and that is exactly what fake accounts cannot reproduce.
There is legal exposure. Posing as a consumer without disclosing a paid relationship may violate Article 21 of Taiwan’s Fair Trade Act. If a competitor reports you, or a consumer digs it up and makes it public, the damage to your brand is worse than never having run word-of-mouth marketing at all.
Five Core Questions for Vetting a Vendor
1. Is the roster a pool of real everyday creators, or a pool of operator accounts?
This is the single most important question, and plenty of vendors will muddy the answer.
A pool of everyday creators: real users with their own day-to-day posting habits, account histories completely unrelated to your brand, recruited through product trials or campaign briefs to share their genuine experience.
A pool of operator accounts: accounts that exist purely to take on commercial work. Their whole reason for existing is to publish sponsored posts, and they have no real personal usage history.
The question to ask is simple: “Before this project, what was the most recent non-sponsored post from the accounts you’ll be using? Can you send screenshots?” Only vendors who can answer that concretely are genuinely working with real people.
2. Is there a fully executed NDA process?
A word-of-mouth strategy contains your marketing direction, competitively sensitive information, and the messaging angles you’re betting on. If the vendor has no proper confidentiality process, your strategy can end up in front of a competitor without you ever knowing.
Ask: “Do the people executing the work sign an NDA before they start? Can I see the template?” If you don’t get a clear answer, or the vendor waves it off with “we have mutual confidentiality,” raise your guard.
3. How is “success” defined?
“Raising brand buzz” is not a definition of success — it’s filler. A reasonable definition should include:
- The platforms in play (which PTT boards, which Dcard forums)
- The number of posts and how they are distributed over time
- Account criteria (account age, historical engagement volume)
- Trackable metrics (reply counts on posts, movement in Google search rankings)
If a vendor is vague about defining success and keeps falling back on “you’ll feel the difference” or “the buzz will pick up,” it usually means they cannot deliver verifiable results.
4. Can you verify the work afterwards?
Legitimate vendors are not afraid of clients checking their work. Once execution is complete, you should be able to:
- Find the corresponding posts directly on the platform
- See that the accounts have a normal history of non-sponsored posts
- See genuine replies or comments on the posts
If the vendor says “we can’t share the accounts” or “checking would interfere with the operation,” that is a huge red flag.
5. Is the pricing structure transparent?
Quotes for word-of-mouth marketing in Taiwan can differ by more than fivefold, and the gap comes mostly from the quality of the posting accounts. Get clear answers on:
- How many people will be doing the posting
- The track-record standard those accounts must meet
- Which platforms and which boards are included
- Whether a performance report is included
When you compare quotes, don’t look only at the total. Work out the “account-quality cost” per post — cheap almost always means the accounts are weak.
Common Contract Traps
A “results after publication are not guaranteed” disclaimer
Within reason, this is acceptable — nobody can control how PTT users react. But if the vendor won’t even guarantee the number of posts published, or bears no obligation at all when results fall short, you have a problem.
Revisions not included in the fee
Word-of-mouth posts need to be tuned to the tone of each platform, and you have to be able to revise a first draft that misses the mark. If the contract bills revisions separately, every rejected first draft costs extra and your budget balloons without limit.
“Monthly billing, cancel any time” — with a three-month minimum spend
That structure usually means the vendor knows the first three months are when clients are most likely to be unhappy, so they lock in the revenue up front. Vendors with real confidence tend to offer far cleaner cancellation terms.
A Suggested Evaluation Process
If you’re evaluating several vendors at once, this sequence is the most efficient.
Round one — paper screening: send every vendor the same brief and ask for “sample posting accounts (screenshots of live accounts)” and “past case studies in a comparable industry.” Anyone who can’t produce them is out.
Round two — video call: walk through the five core questions above one by one. What matters is whether their answers are specific and whether they can back them up with examples.
Round three — small pilot: if a vendor clears the first two rounds and you’re still unsure, ask for a small paid pilot (three to five posts over one to two weeks) so you can verify execution quality before committing to a full engagement.
Taiwan’s word-of-mouth marketing market is in the middle of a cleanup. Vendors genuinely working with real accounts are still in the minority — but the gap in results between them and everyone else is stark.
The single most important criterion when picking a partner: is the vendor willing to let you verify the work afterwards? If they’ll let you check, nine times out of ten they’re doing real work.
If you’d like to see whether NETVANA is a fit for what you need, book a free consultation and we’ll walk you through the actual accounts and materials we use.
Further reading: for the legal exposure that comes with fake-account operations, see our guide to word-of-mouth marketing compliance in Taiwan; to plan a word-of-mouth budget, see the complete guide to word-of-mouth marketing costs; and for how everyday-creator KOC programs are actually run, see the complete KOC seeding playbook.
