"Anyone tried the vitamin C serum?" and "I wish they made it fragrance-free" are both neutral. One is a shopper about to buy. The other is a note for your product team. Mentionfront gives every mention four labels (sentiment, intent, theme and a priority score) so you know which is which without reading them all.
No credit card. 7-day free trial.
Switched from Lumaire to Glowbar last month. Anyone else find it gentler on redness?
The reasons line is shown under every mention in your feed.
A positive/negative label tells you how something was said. It doesn't tell you who should respond or how soon. These four mentions show why Mentionfront adds intent on top of sentiment.
| The mention | Sentiment | Intent | What to do with it |
|---|---|---|---|
| "Anyone tried Glowbar's vitamin C serum? Worth the hype?" | Neutral | Recommendation request | Reply today with a real reason to try it. It lands in Opportunities. |
| "Wish they made the Barrier Cream fragrance-free. Would buy if they did." | Neutral | Feedback | No reply needed. Count it toward the case for a fragrance-free version. |
| "Second time my order arrived damaged. Still no response." | Negative | Complaint | Get support on it before it spreads. It lands in Needs attention. |
| "Loyal since 2022. Glowbar's night balm is genuinely the best." | Positive | Advocacy | Thank them, then save it with Collect testimonial. |
Each intent adds a fixed number of points to the priority score. The more commercially urgent the intent, the more points it adds. Intents are checked in a fixed order, purchase signals first, so "adding it to cart, anyone have a discount code?" counts as ready to buy rather than a question.
The shopper is at checkout or close to it.
"Adding the Barrier Cream to cart. Anyone have a discount code?"Lands in Opportunities
Switching brands, or asking what to use instead.
"Switched from Lumaire last month and honestly never going back."Lands in Opportunities
Something went wrong with an order, a product or support.
"Second time my order arrived damaged. Still no response."Lands in Needs attention
Weighing it up and asking other people first.
"Does the night balm work on combination skin? Should I buy it?"Lands in Opportunities
A hesitation that stops the sale, usually about price or trust.
"Looks lovely but too expensive for a moisturizer. Hesitant."Lands in Needs attention
Unprompted praise and recommending you to others.
"Loyal since 2022. Genuinely the best night balm I've owned."Often a testimonial
Wishes, restock requests and "I'd buy if…" notes.
"Wish they made it fragrance-free. Would buy if they did."Shows in Analytics
Relevant, but nothing to act on: news, sightings, chatter.
"Glowbar pop-up at the mall today, the line was wild."Stays in All
Knowing you got 31 negative mentions this month is only half useful. Knowing that 11 of them were about packaging tells you who to call. Every relevant mention gets one of ten themes, written for ecommerce rather than PR:
On the Analytics page, filter by sentiment to see themes for just the positive or negative mentions. That answers questions like "what are people praising us for?" and "what are the complaints about?" Filter to one product or one competitor tracker when you need a narrower answer. Then filter the inbox by the same sentiment, intent and tracker to read the posts behind the numbers.
Negative mentions by theme
Example workspace · last 30 days · 31 negative mentions
The priority score adds up three things, and each mention lists which ones applied. That's what sorts your inbox, and it's the number your alert threshold is compared against.
30 if the post names a tracked term as a whole word. 12 for a near match, like the plural "Glowbars". Posts that only contain the term inside a longer word aren't collected, and posts with one of your exclude terms are dropped before scoring.
The fixed weight for its intent. Ready to buy adds the most (40) and a complaint adds 28, so both rank above praise (12) however enthusiastic the praise is.
Whether the post names the brand, product or competitor behind the tracker, including its aliases and SKUs. A whole-word match is 90% confidence and adds 27 points. A partial match is 55% and adds 17.
Word lists fail on the way people actually write. "Not bad" contains "bad". "Great, it broke again" contains "great". So sentiment is read by an AI language model, one mention at a time, and it judges the post's tone toward your brand, not how upbeat the wording is.
That matters most for complaints wrapped in jokes. A sarcastic post about a broken pump belongs in the negative pile even if half the words are cheerful. It also matters for plain-sounding praise like "third tub, not switching," which is a clear compliment with no exclamation marks.
Great, the Glowbar pump broke again. Love that for me.
Not bad at all, actually. Third tub of the Barrier Cream and not switching.
Grabbed Glowbar at the airport. It's fine I guess, smells like nothing.