Mona
Makes your ads.
- Static ads and carousels, built around your real product photo.
- Your type, colours and voice. The words are set by code, never drawn by the image model.
- Nothing posts and nothing spends without your click.
Waitlist openFirst brands get in early
Mona makes your ads. Luce tells you if people want what you’re making next. Both learn your brand before they make anything, and keep learning from every yes and no.
Free to join · No card · Nothing posts or spends without your click
Sample brands · not real companies
Two products
Two products. One memory.
Mona makes your ads. Luce finds your next product and tests it before you make it. Each works on its own; use both and they learn from the same memory of your brand.
Luce
Finds your next product, and tests it before you make it.
Drowsy’s next product? Three concepts
Travel$24
Mist$28
Unwind$32
- Travel19$7.89 per deposit
- Mist6$25.00 per deposit
- Unwind11$13.64 per deposit
Luce compares. Whether to make it is your call.
- Finds what people keep asking for on Reddit, X, forums and reviews.
- Shapes it into concepts, each with a look, a promise and a price, then tests them on Meta with your OK.
- Counts real intent, like paid deposits. Whether to make it, and how many, is your call.
Brand memory
Shared. Mona and Luce both read it, and both add to it.
- Dry jokes, never about being tiredSetupyour onboarding
- Cobalt and orange. Never pale, never greyMona6 of 7 approvals
- Can say “melatonin-free”. Can’t say “clinically proven”Setupyour claims sheet
- Next: test Travel at $24 vs $29Luceyour call in Luce, “most deposits”
LOUD, Drowsy and Zest are sample brands. Numbers are sample data.
How it starts
Most AI tools read your website. Solse interviews you.
Reading your store is where it starts, not where it stops. Then it asks what a good creative director and a good product lead would, writes your answers down as rules you can read, and keeps asking as it learns. One setup for Mona and Luce, all in the app. No calls, no salespeople.
- Step 01
It reads your store. That’s the first draft.
Products, photos, reviews, colours, type and the way you write, in about two minutes. Then it shows you what it understood. Cross out what’s wrong; your correction wins.
- Step 02
Then it interviews you, about your brand and what’s next.
By voice or text, in the app. For your ads: what you never say, what you can claim, who it’s for. For what’s next: the products you’re thinking about, and what counts as real interest for you (a waitlist, a deposit, a pre-order). Talking is quicker and catches more. Stop whenever.
- Step 03
Every rule shows why, and where it came from.
Your answers become rules you can read and edit, each with its source. Anything that can’t be a rule is kept as context, labelled, and never quietly enforced.
- Step 04
It keeps learning, and asks first.
Say no and it asks why. It proposes changes to its own rules; you approve them. Small taste tweaks can apply and be undone. Claims, spend and boundaries always wait for your click.
| Most AI tools | Solse | |
|---|---|---|
| Your website | Read once, turned into a brand kit | Read as a first draft that you correct |
| Your brand | Logo, colours and fonts | An interview: what you never say, what you can claim, who it’s for, how to show something new |
| Your next product | A separate tool, or a guess | Same setup: Luce already knows your brand, your customers and what counts as real interest |
| The rules | Hidden in a style or a prompt | A rulebook you can read. Every rule shows why and where it came from |
| When you say no | You re-prompt and try again | It asks why. Taste is learned; a performance claim needs a reason or data |
| Over time | The same setup until you redo it | It proposes changes to its own rules, and you approve them |
Then Mona makes your first ad, or Luce starts a test. Both read the same rulebook.Drowsy is a sample brand. The screens show how the product is designed; it’s in development.Join the waitlist
Mona · makes your ads
Ads of your real product, in your brand’s own look.
Static ads and carousels for Meta and anywhere else you advertise. This is what Mona does, in order.
1
Starts from your rulebook.
Not a template: what you said in your interview and what it read in your store, each rule with its reason.
2
Keeps your real product.
Each scene is built around your own product photo and checked against it. A kind of product it hasn’t made before? It asks how you’d show it first.
3
Sets your words in your type.
The line is live text in your font, spelled right. The image model never draws a letter.
4
Asks why when you say no.
Any answer teaches it taste, even “just a vibe”. A performance claim needs a reason or data. It proposes the rule; you approve it.
5Optional
Suggests. Never acts.
Connect Meta and Mona watches CTR, CPM and cost per purchase, then suggests. You click.
You see the price before anything is made: one ad first, then the batch if you want it. Images that fail are free.
Your product stays your product.
Every ad is compared with your own product photo before you see it. If the lid drifts a shade or the moon loses its shape, it’s flagged.
- Shape
- Lid colour
- Label
- Moon mark
Every ad shows its reasons.
Tap any ad to see why it looks the way it does, quoted from you, your approvals or your reviews.
Why this ad
Ad 04 · Cloud nine
- A dry joke, not a yawn“Funny, never about being tired.”your onboarding
- Cobalt night, orange lidApproved in cobalt every time but once6 of 7 approvals
- Nothing it can’t prove“Melatonin-free” yes, “clinically proven” noyour claims sheet
Drowsy is a sample brand.
Luce · finds and tests your next product
Find your next product. Test it before you make it.
Luce finds what people are asking for, shapes it into product concepts with a look, a promise and a price, and tests them with real money on the line. Here, Drowsy decides what to launch next.
1
Explore the demand.
It reads Reddit, X, forums and reviews for what people keep asking for, quoted and counted. Your brand or a new one.
2
Shape three concepts.
A look, a promise and a price for each, made in your brand. Pick, edit or ask for more.
3
Test them for real.
A page and an ad per concept, the same budget each, on Meta with your OK. It counts paid deposits, not likes.
4
Read the numbers.
Clicks show the ad’s pull, deposits the product’s. Every concept compared, then read out in plain English.
5
You decide.
Whether to make it, and how many, is your call. Luce suggests the next test and remembers what you chose.
You set the budget and how long it runs before anything starts. Nothing spends without your click, and ad spend goes straight to the ad platform.
Three possible products. Clear numbers.
Each concept got its own page and ad and the same budget. Luce lays the numbers side by side and reads them out; you make the call, and it remembers why.
Which product should Drowsy launch next?
Sample test · 14 days · $150 per concept
- Testing3 new products
- Samebudget, ad layout, page
- Counting$5 refundable deposits · set by you
LuceLuce’s read-out
- Travel took the most deposits (19) at the lowest cost per deposit ($7.89).
- Mist drew the most clicks (164, 4.04% CTR) but only 6 deposits: the ad pulled people in; fewer of them put money down.
- Unwind sits in between: 11 deposits at $13.64 each.
Suggested next testTest Travel at $24 vs $29
Your callWhether to make it, and how many, is up to you. Pick one to record your decision.
Sample numbers. Luce compares and suggests tests; the business call is yours.
The ad and the product, measured apart.
Clicks tell you how well the ad pulled. Paid deposits tell you whether people want the product enough to put money down.
- Drowsy Travel$24 · Jet-lag sleep gummies in a slim pocket tin
Land. Sleep. Function.
91 asking - Drowsy Mist$28 · Pillow spray
Two sprays, lights out.
34 asking - Drowsy Unwind$32 · Magnesium night drink, in sachets
A warm cup that switches you off.
22 asking
Clickstell you aboutthe ad’s pull
- Travel1182.80% CTR
- Mist1644.04% CTR
- Unwind1022.44% CTR
Paid depositstell you aboutthe product’s pull
- Travel1916.1% of clicks
- Mist63.7% of clicks
- Unwind1110.8% of clicks
Mist drew the most clicks (164) but 6 deposits; Travel drew fewer clicks and 19 deposits. A good ad isn’t the same as a product people pay for.
Drowsy is a sample brand. Its concepts are Luce concept visuals, and every number here is sample data.
Brand memory · shared by Mona and Luce
Your brand, as a rulebook you can read.
Not a black-box brand kit. Every rule says why it exists and who added it: you, your approvals, your reviews, or Solse with your OK. Mona and Luce read the same rulebook and both write to it. Edit anything, switch anything off.
Teach it
Switch on a rule. Watch the ad change.
These rules apply to every ad Mona makes and every test ad Luce runs for Drowsy.
Clinically proven deep sleep.DrowsyRules on: 0 of 4. Could be anyone’s.
The rulebook
Drowsy · sample- Never “clinically proven”“Legal says no.”Youclaims sheet, setup
- Say “like someone turned the volume down”Customers’ own words for the first weekReviews184 reviews
- Cobalt and orange, never paleApproved in cobalt every time but onceApprovals6 of 7, in Mona
- New products: shot at night, product firstIt asked before making the first Travel tin adYouasked you once, in Mona
- Next: test Travel at $24 vs $29“Most deposits.” Kept with the numbers you sawYouyour call in Luce
Proposed by SolseBoundary · waits for you
Hands only in frame, faces outYou turned down 4 ads with faces: “feels like stock.”
Kept as context, not a rule
“It should feel like a night out, not a pharmacy.”
From your interview. Mona reads it as background; it isn’t enforced and never overrides a rule.
Two layers
The craft is shared. The taste is yours.
- Craft layerLight, composition, type setting, how products sit in a sceneImproves for every brand
- Drowsy’s taste layerThe rulebook, your picks and your reasonsOnly Drowsy
- Your taste in anyone else’s adsNever
Every ad, scored three ways. Never blended.
- Fidelity
- Is it your product? Checked against your photo. Taste can’t override it.
- Fit
- Is it you? Learned from your rulebook and your picks.
- Realism
- Does it pass as a real photo?
What you teach it is never used for another brand. That’s why no two brands end up looking the same.
Pricing
Founding prices. Locked while you stay.
Join the waitlist and the price you see is the price you keep. Nothing is charged until your invite, and there’s no card today.
Starter
75 new ad images a quarter
$19/ monthvs $29 monthly
$57 every 3 months · saves $30 a quarter
- Unlimited versions of each: new lines, sizes, layouts
- Statics and carousels
- Built from your real product photo, checked against it
- Your type, colours and voice in your brand memory
Growth
Recommended300 new ad images a quarter
$39/ monthvs $54 monthly
$117 every 3 months · saves $45 a quarter
- Everything in Starter
- Meta suggestions when you connect it (optional)
- More products and campaigns side by side
Scale
600 new ad images a quarter
$79/ monthvs $119 monthly
$237 every 3 months · saves $120 a quarter
- Everything in Growth
- For brands running new ads every week
- First in the queue when it’s busy
- Prices in US dollars, before any tax. Cancel any time.
- Quarterly plans pool the allowance: use the three months’ worth whenever you like.
- Prices may change for people who join later. The price you see when you join is yours if you subscribe within 30 days of your invite, for as long as you stay. The founding-price rules.
FAQ
Questions.
PermissionsDrowsy · settings · Mona and Luce
- Post an ad without your clickNever
- Spend money without your clickNever
- Decide what you sell, or how muchNever
- Use your photos for another brandNever
Locked. There’s no switch for these.
When can I use it?
Soon. Join the waitlist and we’ll email you when your spot opens.
What will it cost?
Founding prices are in Pricing, above. The price you see when you join is the one you keep. Nothing is charged until your invite, and you can cancel any time.
Isn’t this what every AI ad tool does?
Most read your website and hand you a brand kit. Ours reads it too, as a first draft you correct. Then it interviews you like a creative director would, writes your answers down as rules you can read, each with its reason and source, and asks why when you say no. Over time it proposes changes to its own rules, and you approve them. That’s how it’s designed; it’s in development now.
Do I need both?
No. Mona is an ad maker for your everyday ads. Luce finds and tests new products, with Mona’s ad engine built in for its tests, so it doesn’t need a Mona plan. Use either one, or both: they share one brand memory.
Do I have to get on a call?
No. There are no salespeople and no calls. The interview happens in the app, by voice or text, whenever suits you. Talking is quicker and catches more, and you can stop and pick it up later.
What if I don’t like what it learned?
Open the rulebook and change it. Every rule shows why it exists and where it came from, and you can edit it or switch it off. Small taste changes can be undone; anything about claims, spend or boundaries never changes without your click. Your decision is always final.
Will it post or spend money on its own?
No. In Mona or in Luce, nothing goes live and nothing is charged without your click.
Will Luce tell me which version to make?
It shows which version people responded to, with clear numbers, and can suggest what to test next. Whether to make it, and how much, is always your call.
Is this made with AI?
The scenes are made by image models. Mona sets the text, checks your product in every image, and keeps the record of what you approved. Every image carries an invisible marker saying it was made with AI, as EU law requires.
What happens to my product photos?
We use them to make your ads and test ads. They’re never used for another brand.
What kind of products does it work for?
Physical products with a photo: supplements, skincare, sauces, candles, bags, jewellery and the like. Mona makes static ads and carousels, no video yet, because it isn’t good enough. Luce finds and tests new products before you make them: for a new brand, the next SKU, or a new flavour, size, price or promise.
Join the waitlist.
We’re opening to a small number of brands first. Tell us where to find you.