On X, shoppers don't write reviews. They write "@yourbrand where is my order", "site crashed at checkout" and "anyone have a code?". Mentionfront collects posts and replies that name your brand, products and competitors, separates complaints from buying questions, and puts the urgent ones first.
No credit card. No X API plan to buy.
@fernwickhome ordered 12 days ago, tracking hasn't moved since it left the warehouse. 3 emails, no reply.
Same thing happened with my Fernwick throw. Support finally answered on day 9.
One public complaint, then a second customer confirming it in the replies.
X skews toward reactions in the moment. For an online store that means a different mix than TikTok or Reddit: fewer reviews, more problems and more timing-sensitive posts.
Customers tag the brand after email support goes quiet. Each one is public and searchable, and the replies under it often turn into "same here". Replying fast, then moving the details to DMs, stops one late parcel from becoming a thread.
Sellouts, checkout errors, bots, "was it even in stock?". Limited drops create a burst of posts in a few hours, and the tone of that burst shapes how the next drop is received.
"Anyone have a code for…" posts are buying intent. So are the replies sharing codes, which is how influencer and one-time codes end up far outside their intended audience.
A price rise, a new shipping fee or a changed returns policy gets discussed on X quickly. Early posts tell you whether customers are shrugging or organizing.
Not every mention on X needs your brand account. This is how most small ecommerce teams split the work, using the labels each mention already has.
| Post | Label | Reply publicly? | Then |
|---|---|---|---|
| Late, lost or damaged order | Complaint · shipping & delivery | Yes, one line saying you've seen it | Move to DM for the order number. Resolve in Mentionfront once it's sorted. |
| Question before buying ("does it ship to Canada?") | Recommendation request | Yes, answer in full | If it's a common question, add it to the FAQ or product page. |
| Asking for a discount code | Ready to buy | Only with a public offer | Point to a sitewide promotion if one's running. Don't hand out private codes in replies. |
| Switching from a competitor | Competitor alternative | Yes, if they're asking for suggestions | Answer the reason they're leaving. Don't criticize the other brand. |
| Joke or meme that names you | Other · usually low priority | Usually not | Leave it unless it's taking off and your brand voice suits a reply. |
| Angry post about a policy change | Objection or complaint | Once, with facts | Share the pattern with whoever owns the policy. Don't argue in replies. |
A creator's 30%-off code or a single-use win-back code gets posted to X, picked up by deal accounts and coupon sites, and suddenly shows up on orders from people who would have paid full price. The margin hit only shows up when someone checks discount reports.
Two keywords catch a lot of it:
When a code you meant for one audience is circulating, you can cap its uses or retire it, and check the affiliate attribution on orders that used it.
🔥 Fernwick code LAURENLIVES30 still working for 30% off everything, stack with free shipping
the Fernwick code from that deals account still works, just saved $60 on a duvet
Codes rarely leak quietly, and they rarely leak slowly. Both of these are public posts from ecommerce subreddits.
We use affiliate marketing for our online store, but do currently not give out any discount codes anymore. The reason is that in the past, those codes would leak immediately to coupon sites.
we just ran a 20% auto-discount for a flash sale and i didn't realize an old affiliate code from months ago (15% off) had the "combines with product discounts" box checked. someone found it, posted it on a deal discord, and we got hammered for about 3 hours before i woke up. sold a bunch of units way below cost.
someone posted it on a coupon forum and it went viral. we sell furniture and people were buying $2,000 sofas for $20. i've got 6,300 orders sitting in shopify right now…
The third one was a typo, not a leak — 99% off instead of 9% — but it shows the speed: three days live, spotted by a coupon forum, thousands of orders. Tracking "[brand] code" won't stop a mistake like that, but it puts the post that spread it in front of you the same day, instead of in a discount report at month end.
Mentionfront checks X on a schedule, not second by second. That makes it the right tool for the day before and the week after a launch, and the wrong one for watching checkout minute by minute. Here's how to split it.
Add the product or collection name a few days ahead. Pre-launch posts show which details people are unsure about (sizing, price, ship dates) while there's still time to answer them in your launch email.
For the first hour of a drop, keep X's own search open on the Latest tab for your brand and product name. Nothing beats it for seeing a checkout error the minute it happens.
Over the next few days, sort the launch keyword by intent. How many complaints were about stock versus the site? How many people asked about a restock? That's your brief for the next drop.
Not collected:
Why there's no X API bill
X's own API is priced per record read: its developer docs list post reads at $0.005 each, capped at 3 million a month, so pulling 100,000 posts about your brand is a $500 line item before anyone has read one. Mentionfront collects X posts through commercial search access included in your plan, so a busy brand doesn't mean a surprise bill.
Track your handle and your name
X posts tag "@fernwickhome" far more often than they write "Fernwick". Track both as separate keywords, plus any short name customers use.