Achadu finds and verifies deals, understands shopper requests, prepares and distributes posts, and attributes results to their source. Five connected stages form one complete workflow.
Modules share the same context. What Shopper listens to influences the curation; what sells improves the next recommendation; what fails returns as an alert to the operator.
Monitoring, price history, offer confidence and compatible coupons before publication.
You say the words of your niche, the price range, the minimum commission and how often to look. Achadu monitors Shopee (official API) and Mercado Livre and places what passes the filter in your showcase, as pending or already approved.
Everything that comes in goes through the storefront: import via link, search at the source, register by hand. You approve, reject, see the preview exactly as it will arrive in the group and publish it in one click.
Before recommending, Achadu brings together signals that help separate real discount from promotion just in title: price and availability history, origin reputation, link quality and product consistency between stores.
Register coupons with the rules that actually exist: fixed or percentage discount, minimum order, ceiling, category and validity. For each offer, Achadu calculates which ones are compatible and orders them by estimated savings.
Shopper by text or photo, requests without results, own website and links marked with your commission.
On your website, the person describes what they are looking for or sends a photo. Shopper identifies the item, matches intent with budget and preferences, and searches only among the offers you've published — always isolated in your operation.
When there isn’t a good option, the conversation doesn’t end with “I didn’t find it”. It becomes a prioritized request on your dashboard. New publications are compared with the queue and a compatible combination becomes an opportunity for the operator.
Paste a link or an entire message. Leave with your identifier in each link, without third-party trackers. In Shopee with API, the link becomes short. If an ID is missing, Achadu warns you before publishing.
Set up your offer store, blog, link page and campaigns with your brand. You work on a draft, publish it when it's ready and you can go back to a previous version without dismantling what was already live.
Copy, creatives, segments, agenda, pilot and automations working in the same flow.
The AI writes the copy based on the title, price, discount and coupon: two or three lines, concrete, in the tone of someone recommending it to a friend. No promise of health, no "unmissable", no dash.
Good morning, notice, coupon of the day or offer: once, every day, every week or every N hours. And the automatic pilot by segment: at each interval, within the window you choose, the best approved offer comes out alone.
A segment is your niche: it contains its groups, rules, and autopilot. Publishing groups receive deals; monitoring groups track joins and exits. One number can serve multiple segments, and the public site gets one page per segment.
Choose a deal and its Story creative is generated with the photo, original price, discount, coupon, and your brand in six 1080×1920 templates. Creative Studio also includes your /bio: an editable mini-store built in the same site editor.
A group of your own with the connected number. Send /publicar <link> #segmento, /converter <link>, /cupom shopee CODIGO or /status and Achadu responds there, without opening the platform. Contingency: if number 1 falls, 2 takes over.
Origin of the commission, group behavior, relationships between products and recommendations with visible reasons.
Achadu crosses commission, observed conversion, offer confidence, affinity with the group, time and recent repetition to suggest the next send. The recommendation shows the reasons and separates evidence from hypothesis.
Shipping, click, group, product and commission are linked in the same history. Repeated imports do not double revenue and discrepancies between what was expected and what was received appear for investigation.
Sends by day with period, segment, group and state filter, and export in CSV. Group analysis: entries, exits, balance and engagement by group. Commissions per store, with Shopee synchronized.
The connected number counts the volume of messages from each group, day by day, without reading anyone's content. Crossed with inputs and outputs, Achadu calculates the correlation (too much noise scares people away) and the AI reads the numbers and returns a concrete retention action. There is also the evasion report: who entered and left within 24 hours.
The graph connects products that work together, serve as accessories or can replace one another. For recurring items, you confirm the repurchase cycle and Achadu shows the next windows when there is a registered purchase.
Campaigns, smart invites, capture pages, and full-funnel fraud protection.
Capture page ready by niche (/vip/seu-nicho), with Meta and Google pixel, Lead triggered by the server via Conversions API and the asset that no advertisement has: the connected number sees who ENTERED the group. The funnel shows clicks × actual entries, head by head.
Every click on your invitation goes through anti-fraud before counting: robot, crawler and repeated click are redirected normally, but do not enter the funnel and do not trigger the pixel. The Lead who trains your Meta campaign is just real people.
The dashboard shows the evidence, the level of uncertainty and the decisions that still depend on the operator.
Before recommending, the system needs to know what it is looking at.
Recognize the same product in offers from different stores without combining items just because the name sounds the same.
Summarizes price, availability, reputation, and link quality in an explained note.
Remembers budget and preferences with pseudonymous, revocable and isolated identity per store.
Records compatibility, accessories, substitutes and repurchase cycles with auditable confidence and evidence.
More than ordering by discount: understanding audience, moment and return.
Learn categories and price ranges that work for each audience, always showing the sample size.
Flags trends by category and segment to prepare curation before the peak.
Compare stores selling the same product and consider price, commission, trust and observed performance.
Separate coincidence from real change before escalating a decision.
It compares treated and control groups and makes it clear when the data do not yet allow us to confirm the cause.
It shows when a publication helped or took away space from another, avoiding confusing exchange with growth.
Test price ranges in a controlled way and measure incremental gain, without changing everything at once.
Anticipate problems, explain deviations and preserve purchase intention.
Look for dips and divergences in the journey and point out hypotheses with evidence, even when the sample is small.
Simulates the grid before changing and compares suggestions with human decisions; Never publish alone in this mode.
Reroutes failures, uses contingency and preserves the order when an offer disappears or changes price.