N8n Developer Needed: AI-Powered Business Performance Analyzer

Jobs
1

Amazon Licensing Business — Weekly Intelligence System

The Job

Build an automated n8n system that pulls our Amazon Ads data, product catalog (stored in ClickUp), and BSR tracking (from Amazon.com), analyzes it with Claude AI, and delivers weekly actionable reports to three team members via Slack.

Platform: n8n Cloud Pro
Timeline: 3 weeks
Budget: $1,500 — $3,000 USD (fixed price preferred, phased payment)

The Business

We run an Amazon Merch on Demand business from the UK but operate in the US, that has 10’s of thousands of products in our catalog ~7,400 live products. Our ad spend is 48.3% of revenue against a 40% target — that gap costs us ~$37K/year. We need a system that finds the waste and opportunities we’re missing.

Key numbers:

  • $219K/year ad spend across thousands of campaigns

  • $2.67 average royalty per sale (at $16.99 price point)

  • $3.12 average ad cost per order (ads lose money without organic lift)

  • 595 products drive 80% of sales — the rest need pruning or repricing, we need a way to track the products that are really driving revenue for us and to cut the others

Key constraint: Amazon Merch has no Selling Partner API. However, the Amazon Advertising API IS available, and Keepa provides BSR/price tracking for Amazon.com + we have our product catalog stored in ClickUp so it should be possible to create a system that can check our catalog in ClickUp > Amazon.com and Amazon Ads.

What You’ll Build

3 Sub-Workflows → 3 Slack Channels

1. Ad Waste Detector → Slack #ads-actions (for our PPC manager)

  • Daily: Flag critical waste — campaigns with 10+ clicks / 0 orders, ACoS > 40%

  • Weekly: Deep analysis — total wasted spend breakdown, budget pacing, scaling opportunities, Top 5 actions ranked by dollar impact

  • Data source: Amazon Advertising API

2. Pricing Optimizer → Slack #listings-tasks (for our listings manager)

  • Weekly: Scan catalog for underpriced bestsellers and overpriced dead stock

  • Flag designs at $13.38 ($0 royalty) that have BSR traction and should be $16.99

  • Flag market leaders at $16.99 that should be $19.99

  • Include holiday price-lock warnings (Amazon can lock low prices into deals)

  • Data sources: Keepa API + Amazon Ads API + ClickUp API

3. Performance Tracker → Slack #ceo-brief (for me)

  • Weekly Monday morning brief: One-sentence health verdict, key metrics vs last week, what’s working, what’s bleeding, top 3 actions this week

  • AI-generated strategic analysis using Claude Sonnet

  • Data sources: All of the above + weekly royalty CSV upload

Tech Stack

Component Details
n8n Cloud Pro ~10K executions/month (we’ll use ~300)
Amazon Advertising API OAuth 2.0. Campaign/ad group/keyword performance data
Keepa API API key. BSR history and price tracking per ASIN (€19/mo plan)
ClickUp API API key. Our product catalog — ASINs, niches, tags, status
Claude API (Anthropic) Haiku for daily scans, Sonnet for weekly analysis
Supabase Free tier. Store weekly metrics for week-over-week comparison
Slack Free plan. 3 channels with Block Kit formatted messages (We do have a paid plan currently for our team)

We provide all API credentials and account access.

Deliverables & Payment

Phase Week Deliverable Payment
1 Week 1 Ad Waste Detector (daily scan + Slack alerts). Amazon Ads API integrated, Supabase tables created, basic Slack output working. 33%
2 Week 2 Weekly deep analysis + Pricing Optimizer. Keepa + ClickUp integrated, pricing recommendations with estimated revenue impact, polished Slack formatting. 33%
3 Week 3 Performance Tracker + documentation. Full Monday morning CEO brief, error handling, Loom walkthrough, handover call. 34%

Required Experience

  • n8n: 5+ production workflows with sub-workflows, scheduling, error handling

  • REST APIs: OAuth 2.0 (Amazon), API keys, pagination, rate limiting

  • Claude or OpenAI API: Structured prompts and parsed output in n8n

  • Supabase/PostgreSQL: Basic table design via n8n nodes

  • Slack Block Kit: Formatted messages with sections and fields

Bonus: Amazon Ads API experience, Keepa API, e-commerce background

What’s Next After This

There is a Phase 2 project (Trend Hunting System — larger scope, $2,500-5,000) planned once this system proves ROI. Same developer gets first refusal. Full architecture document available.

To Apply

Send:

  1. 2-3 examples of n8n workflows you’ve built (screenshots, Loom, or exported JSON)

  2. Your experience with LLM APIs in n8n — show an example if possible

  3. Timeline and fixed price for Phases 1-3

  4. One question or suggestion about this brief (shows you’ve actually read it)

Reference “Business Performance Analyzer” in your application.

Thanks in advance,

Colin H.

4 Likes
2

If you need i am available here:

I am one of top 100 verified template creators:

I am also capable of building custom community n8n node too:

  1. https://www.npmjs.com/package/n8n-nodes-pdfbro
  2. https://www.npmjs.com/package/n8n-nodes-ocrbro
  3. https://www.npmjs.com/package/n8n-nodes-ttsbro

Apart from that I’m also a full-stack developer with the right Gen AI experience, which makes me a solid plus for your team[but right now only vibecoding]

Check my recent gen ai projects… I built a native Android automation agent too. It’s worth a look:

I can build complex AI automations directly in code, not just inside n8n

I recently started posting my n8n work on YouTube with explanations:
https://youtube.com/@blankarray

You can schedule a quick call with me: https://cal.com/abhi.vaar/n8n

Fun fact… I even made an n8n workflow to find a few n8n project leads for myself so I truly believe in what I do…

9bSkKu
9bSkKu1220×1470 122 KB

I started asking My recent clients for honest feedbacks so here is one testimonial: https://www.youtube.com/watch?v=TqBy3SVCHgQ&list=PLAJltY5bp6yiZ3sFBjm7bfrkLXSGtJX8m

Here is my linktree: iamvaar | Linktree

And I also built low latency voice appointment scheduler with live ai avatar[in code]:

I built AI search visibility tracker[I am capable of building complex web scraping automations in python too]:

3

Hi Colin,

As an automation engineer with 5+ years of Python and 2+ years building advanced n8n workflows, I’ve helped businesses automate complex reporting and analytics across multiple APIs, including Slack integrations and LLM-powered insights. While I haven’t built a system with this exact Amazon Ads/Keepa/ClickUp stack before, I am sure I can create something similar with my background.

As required, in my website you can find my previous jobs, including workflow blueprints: caiov.dev

As for the timeline and prices, I’d require 4 weeks to deliver the MVP (1 more week for the final step as I’ll need to perform multiple tests to ensure everything is working seamlessly). I agree with the percentages for each phase, but I’d charge USD 2000 for the whole project

If you think it’s a match, please contact me at [email protected] so we can discuss more details further.

Best regards,

Caio Valadares

Automation Engineer | n8n | Python | SQL | Javascript

Website: caiov.dev

Email: [email protected]

Upwork: View my profile

Book a Meeting: Schedule here

5

@chfcm Hi Colin,

Reference: Business Performance Analyzer

I’ve gone through your brief carefully, and this is absolutely something we can build properly.

We’ve developed multiple AI-powered automation systems using n8n that combine API integrations, structured data storage, LLM analysis, and Slack reporting — built for real operational use, not just dashboards.

Your setup with Amazon Ads, Keepa, ClickUp, Supabase, and Claude makes sense architecturally. The key will be structuring the data cleanly so weekly insights are consistent and actually actionable for each role.

You can review our recent projects and workflow breakdowns here:
:link: https://muhammad-ai-automations.notion.site/Muhammad-Bin-Zohaib-AI-Automation-Projects-29da292a241380f889c2e337a134c010

We’re a full AI development and automation agency, and we’re ready to build this end-to-end with proper structure and documentation.

One quick question before moving forward:
For the weekly CEO brief, do you want the analysis focused more on ad efficiency metrics, or on net profit impact per SKU after royalties and ad cost?

Happy to discuss everything in detail and align on the architecture.

You can reach me directly:

Email: [email protected]
Phone / WhatsApp: +92 3360327970

Let’s connect and go from there.

— Muhammad Bin Zohaib

6

Will you treat “595 products drive 80% of sales” as the core cohort and run everything else as an exception/pruning pipeline?

I’ve built production n8n systems where multiple APIs + a warehouse layer feed daily/weekly agentic analysis with strict schemas, idempotency, and Slack Block Kit outputs. A relevant example of our automation discipline is our ESG & Sustainability Automation Engine (multi source ingestion, validation, auditability): Full Case Study

We’ve also shipped workflow heavy automations with clear logging and handover: Walkthrough

For your build, I’d structure it exactly as you outlined: three sub-workflows with Supabase as the week-over-week truth. Phase 1 is your Ad Waste Detector running daily off Amazon Ads API, storing campaign/ad group/keyword rollups, then generating a deterministic action list (10+ clicks/0 orders, ACoS > 40%, budget pacing) before posting to #ads-actions.

Phase 2 adds the Pricing Optimizer by joining ClickUp catalog ASINs + Keepa BSR/price history + Ads performance so we can flag “BSR traction but $0 royalty pricing” and “dead stock overpriced,” with estimated impact and price-lock warnings. Phase 3 is the CEO Monday brief: a single health verdict plus deltas vs last week from Supabase, then Claude Sonnet produces strategic narrative on top of validated metrics, never as the source of truth.

To keep this safe and cheap in n8n Cloud Pro, I’d use strict JSON outputs from Claude (Haiku for daily classification, Sonnet for weekly narrative), caching for repeated ASIN lookups, pagination/rate-limit guards for Amazon/Keepa, and a dead-letter queue so one failing ASIN never blocks the whole run. Slack messages will be Block Kit, formatted per channel, with top actions ranked by $ impact and links back to the underlying campaign/ASIN.

Timeline and fixed price: I can match your 3-week phased plan and propose a fixed bid within your $1.5k–$3k range after a 20-minute kickoff to confirm Amazon Ads scopes, ClickUp catalog schema, and the royalty CSV format.

One question before we lock Phase 1: do you want “waste” calculated at campaign level only, or should we drill down to search term / keyword where available (often where the fastest savings are)?

:star: Reviews: Clutch
:date: Book a call: Calendly

7

I’ve reviewed your brief for the Amazon Licensing Weekly Intelligence System. Given my background as an AI Automation Architect and my deep experience with production-grade n8n environments, I am confident I can bridge the gap between your $219K/year ad spend and your 40% ACoS target.

​My approach centers on reliable data “pipes” and structured AI reasoning. I don’t just send Slack alerts; I build decision-support systems that highlight dollar-impact opportunities.

​Why I’m the Right Technical Partner

  • ​Advanced n8n Architecture: I am highly proficient in building modular workflows with error-handling subflows, parent/child execution, and complex custom JavaScript nodes for data transformation. I currently manage a local Docker environment for high-performance automation.

  • ​API & OAuth Mastery: I have extensive experience with OAuth 2.0 (required for Amazon Advertising) and integrating complex REST APIs like ClickUp and Keepa. I understand how to handle pagination and rate limits to ensure 100% data integrity.

  • ​Claude/LLM Integration: I specialize in using Claude 3.5 Sonnet for deep strategic analysis. I focus on “Structured Prompting” to ensure the AI returns clean JSON that can be parsed directly into Slack Block Kit or Supabase.

  • ​Infrastructure: I am a frequent user of Supabase for persistent storage (tracking week-over-week BSR/Price changes) and Slack Block Kit for highly readable, actionable reports.

​My Proposal (Phases 1-3)

  • ​Timeline: 3 Weeks (as requested).

  • ​Fixed Price: $2,800 USD (Includes comprehensive documentation and a Loom walkthrough for your team).

  • ​Portfolio: You can view my recent automation projects and architectural blueprints here: myportoliowork.lovable.app

​One Strategic Suggestion

​Regarding your Pricing Optimizer: Since Amazon can “price-lock” during deals, I suggest we implement a “Pre-Lock Safety Check” node. This would cross-reference your ClickUp “Status” or “Holiday” tags against the Keepa price history before suggesting a price increase, preventing a situation where you’re locked into a low margin during a peak traffic window.

​Proof of Work

​I have built several complex systems including:

  1. ​LeadGuard AI: A multi-tenant SaaS for lead qualification using Groq and Supabase.

  2. ​The Autonomous Literary Architect: A multi-step pipeline for content enrichment and automated asset delivery. I can provide exported JSON or a video walkthrough of these logic-heavy flows upon request.

​I am ready to help you reclaim that $37K/year in wasted ad spend. When would be a good time to discuss the Amazon Advertising API credentials and the ClickUp hierarchy?

​Best regards,

​Agbaje Olarewaju

AI Automation Specialist

8

Hi there, I’m interested in taking up this project and have sent you a message with all the relevant details. Looking forward to your response.

9

Hi Colin,

This is a clean project with clear deliverables — happy to see the phased approach.

A few thoughts on the architecture:

Amazon Ads → n8n: The SP-API (Sponsored Products, Brands, Display) has a reporting endpoint that works well with n8n’s HTTP Request node. The key gotcha is token refresh — Amazon’s LWA tokens expire every hour, so building a credential rotation subflow upfront saves you from silent failures on week 2. We run this pattern for e-commerce clients already.

BSR Tracking: Since Amazon doesn’t have a public BSR API, this typically needs periodic scraping (Apify or a custom Lambda). In n8n Cloud Pro, you’re limited to HTTP-based approaches — Keepa’s API is the cleanest option here if the product count is manageable (<500 ASINs). Otherwise Jungle Scout’s API works too.

Claude AI Analysis: For weekly intelligence reports, the prompting strategy matters more than the model choice. We structure it as: (1) raw data normalization → (2) trend extraction subflow → (3) Claude summarization with business-specific context. This avoids the “generic AI summary” problem and gives your team actual decisions to make.

Slack delivery: Native n8n Slack node handles this cleanly. We can template the reports with rich formatting — charts as image attachments, key metrics bolded, action items as bullet points.

We run 40+ n8n workflows in production including e-commerce analytics pipelines and Claude-powered reporting. Happy to share a relevant workflow screenshot in DM. Would a quick scoping call work to nail down the API access details?

— Derek (Click Consultants)

10

Hi Colin,

I’ve worked on very similar n8n intelligence workflows that connect advertising APIs, product databases, and AI analysis to generate actionable Slack reports, so your Business Performance Analyzer is very clear to implement.

Based on your brief, I would structure it as three modular sub-workflows with Supabase as the central metrics layer:

Phase 1 - Ad Waste Detector

Integrate the Amazon Advertising API with scheduled daily scans to detect wasted spend (10+ clicks / 0 orders, ACoS > 40%), store performance snapshots in Supabase, and send structured Slack alerts to the PPC channel. This also creates the historical base needed for deeper weekly analysis.

Phase 2 - Pricing Optimizer

Connect ClickUp and Keepa to evaluate your full catalog against BSR trends, pricing position, and ad performance. The workflow will identify underpriced winners, overpriced dead stock, and revenue-impact opportunities, then deliver listing action reports via Slack.

Phase 3 - Performance Tracker

Build the Monday CEO brief by combining weekly metrics, royalty CSV data, and trend comparisons from Supabase. Claude Sonnet will generate the strategic summary, including health verdict, risks, and top priority actions.

The workflows will include proper error handling, pagination handling, and clean, readable structure so your team can maintain them easily.

Suggestion:
One improvement I recommend is storing weekly ASIN-level aggregates in Supabase. This allows Claude to detect performance shifts over time instead of analyzing only single-week snapshots, which makes the recommendations much more accurate.

Question:
Do you already have the Amazon Advertising API developer account approved, or should that be included in the setup phase?

I can start immediately and deliver Phase 1 within the first week.

My mail address “[email protected]”

11

Reference: Business Performance Analyzer

Hey Colin,

I’ve sent you all the details over DM. Please check.

Thank you!

Rohan

12

Hi Colin,

This is a clean project with clear deliverables — happy to see the phased approach.

A few thoughts on the architecture:

Amazon Ads → n8n: The SP-API (Sponsored Products, Brands, Display) has a reporting endpoint that works well with n8n’s HTTP Request node. The key gotcha is token refresh — Amazon’s LWA tokens expire every hour, so building a credential rotation subflow upfront saves you from silent failures on week 2. We run this pattern for e-commerce clients already.

BSR Tracking: Since Amazon doesn’t have a public BSR API, this typically needs periodic scraping (Apify or a custom Lambda). In n8n Cloud Pro, you’re limited to HTTP-based approaches — Keepa’s API is the cleanest option here if the product count is manageable (<500 ASINs). Otherwise Jungle Scout’s API works too.

Claude AI Analysis: For weekly intelligence reports, the prompting strategy matters more than the model choice. We structure it as: (1) raw data normalization → (2) trend extraction subflow → (3) Claude summarization with business-specific context. This avoids the “generic AI summary” problem and gives your team actual decisions to make.

Slack delivery: Native n8n Slack node handles this cleanly. We can template the reports with rich formatting — charts as image attachments, key metrics bolded, action items as bullet points.

We run 40+ n8n workflows in production including e-commerce analytics pipelines and Claude-powered reporting. Happy to share a relevant workflow screenshot in DM. Would a quick scoping call work to nail down the API access details?

— Derek (Click Consultants)

13

Hey Profile - chfcm - n8n Community

I got you, I have been building all forms of automations for the past 2 years and have built 100s of flows for my clients. Have worked with all sorts of companies and gotten them 10s of thousands in revenue or savings by strategic flows. When you decide to work with me, not only will I build this flow out, but also give you a free consultation like I have for all my clients that led to these revenue jumps.

I have built a similar workflow like this for one of my clients. I can not only share that but also how you can streamline processes in your company for faster operations. All this with no strings attached on our first call.

Here, have a look at my website and you can book a call with me there!

Talk soon!

14

Hey Colin,

Just read through your brief — this is a really well-thought-out project. The $37K/year leak from ad waste is painful but totally fixable with the right automation.

I’ve built similar systems for Amazon sellers before, so a lot of what you’re describing is familiar territory for me. Here’s how I’d approach it:

Week 1: Ad Waste Detector

First thing I’ll do is get the Amazon Advertising API connected and pulling your campaign data. I’ll set up Supabase tables to track performance over time so we can spot trends, not just snapshots.

The daily workflow will scan for the obvious bleeding — campaigns burning through clicks with zero conversions, anything over 40% ACoS. Pretty straightforward logic there.

For the weekly deep dive, I’ll use Claude Haiku to analyze the full dataset and rank your top opportunities by dollar impact. The Slack messages will use Block Kit so they’re actually readable (not just walls of text).

Week 2: Pricing Optimizer

This is where it gets interesting. I’ll connect Keepa to pull BSR data and cross-reference it with your ClickUp catalog and ad performance.

The logic here is: if something’s selling well at $13.38 (essentially giving away margin), flag it for a price bump. If you’ve got market leaders still at $16.99 when competitors are at $19.99, that’s leaving money on the table.

I’ll also build in holiday warnings since Amazon loves locking in low prices for deals — better to catch that early.

Week 3: CEO Brief + Handover

The Monday morning brief is basically tying everything together. I’ll use Claude Sonnet here since it’s better at strategic thinking than Haiku.

Format will be simple:

  • One-line verdict (green/yellow/red)

  • Key numbers vs last week

  • What’s working / what’s bleeding

  • Top 3 moves for the week

I’ll also record a Loom walkthrough showing you how everything works, plus we’ll do a proper handover call so you’re not left guessing.

My Background

I’ve been building n8n workflows for about 3 years now, mostly for e-commerce and B2B automation. I’m comfortable with the whole stack you mentioned — Amazon APIs, Keepa, ClickUp, Claude, Supabase, Slack Block Kit.

Here’s a workflow I built recently for a similar project: [Google Drive link to screenshot]

It’s a different business, but shows how I structure sub-workflows, handle errors, and format Slack reports.

Pricing

I’m thinking $2,400 for the full build, split across the three weeks:

  • Week 1: $800 after Ad Waste Detector is live

  • Week 2: $800 after Pricing Optimizer is working

  • Week 3: $800 on final delivery + handover

That includes documentation, the Loom walkthrough, and two weeks of support after launch in case anything breaks.

One Question

You mentioned 595 products driving 80% of sales. Would it be useful to auto-tag those in ClickUp as “core performers” and get alerts when they start declining? That way you’d know immediately if a winner starts bleeding, rather than catching it in the weekly report.

Here are some n8n workflows I’ve built recently:

Shows my approach to multi-agent systems, error handling, OCR extraction, and Telegram alerting. Happy to walk through the architecture if helpful.

Invoice Processing Agent — Email monitoring → AI OCR extraction (Arabic + English) → Validation → ERPNext Purchase Invoice creation
Invoice Processing Agent — Email monitoring → AI OCR extraction (Arabic + English) → Validation → ERPNext Purchase Invoice creation1920×1032 154 KB

Bank Reconciliation Agent — Daily automated matching using multi-criteria confidence scoring (amount, reference, vendor, date)
Bank Reconciliation Agent — Daily automated matching using multi-criteria confidence scoring (amount, reference, vendor, date)1920×1032 156 KB

Financial Reporting Agent — Weekly summaries + monthly AI-powered analysis → Email reports + Google Sheets dashboard
Financial Reporting Agent — Weekly summaries + monthly AI-powered analysis → Email reports + Google Sheets dashboard1920×1032 173 KB

Financial Reporting Agent — Weekly summaries + monthly AI-powered analysis → Email reports + Google Sheets dashboard
Financial Reporting Agent — Weekly summaries + monthly AI-powered analysis → Email reports + Google Sheets dashboard1920×1032 173 KB

  1. Agent 1: Invoice processing with AI OCR (Arabic + English)
  2. Agent 2: Bank reconciliation with confidence scoring
  3. Agent 3: Financial reporting with AI analysis
  4. Agent 4: System monitoring with Telegram alerts

Phase 2

Definitely interested in the Trend Hunting System once this proves ROI. Would love to see the architecture doc when we get there.

Timeline

I can start March 1st as you mentioned. Should have everything wrapped by March 21st.

Let me know if you want to hop on a quick call to go over API access and any questions you have.

Cheers,
Hamza
[email protected]

P.S. - Reference: Business Performance Analyzer

1 Like
15

Hi Colin, this is right in my lane. I build production n8n systems that combine multiple APIs, data storage, and AI analysis into reliable daily and weekly decision workflows. I have shipped pipelines with sub-workflows, scheduling, retries, logging, and Slack reporting, including LLM-based analysis and structured outputs.

Your split into Ad Waste, Pricing, and a CEO brief makes sense and maps well to how I would design this in n8n. I’ve sent you a DM with 2 to 3 concrete examples of similar systems I’ve built, plus a proposed approach and timeline.

Reference: Business Performance Analyzer

16

Hi Colin,

**Re: Business Performance Analyzer**

This is one of the best-structured briefs I’ve seen — you clearly understand both the business problem and the technical requirements. I’m reaching out because my daily work sits exactly at the intersection of **n8n workflow architecture + Claude AI integration**.

-–

### Why I’m a Strong Fit

- **AI Consultant** building production n8n workflows with sub-workflows, scheduling, and error handling — not simple linear automations, but multi-branch systems with retry logic and failure notifications

- **Deep Claude API experience in n8n** — I work with both Haiku (for fast classification/filtering tasks) and Sonnet (for analytical reasoning) daily. I build structured prompts that return parseable JSON, handle token limits, and chain multiple AI calls for complex analysis

- **REST API orchestration**: OAuth 2.0 flows, pagination handling, rate limiting, and credential management across multiple services in single workflows

- **Slack Block Kit** reporting — I’ve built formatted intelligence reports with sections, fields, actionable buttons, and conditional formatting based on severity

- **Supabase/PostgreSQL** table design and querying via n8n nodes for storing time-series metrics and running week-over-week comparisons

-–

### Timeline & Fixed Price

| Phase | Week | Deliverable | Payment |

|-------|------|-------------|---------|

| 1 | Week 1 | Ad Waste Detector — Amazon Ads API integrated, daily scans with Supabase snapshots, Slack #ads-actions alerts working | 33% |

| 2 | Week 2 | Pricing Optimizer — Keepa + ClickUp integrated, catalog analysis with pricing recommendations, polished Slack output | 33% |

| 3 | Week 3 | CEO Brief + documentation — Monday morning intelligence report via Claude Sonnet, error handling hardened, Loom walkthrough + handover call | 34% |

**Total: $1,500 USD fixed price.** I’m pricing at the efficient end because this project maps directly to workflows I’ve already built, and I want to be the right partner for the Phase 2 Trend Hunting System.

Can start immediately.

-–

### Suggestion

For the Ad Waste Detector daily scans, I’d recommend implementing a **severity-tiered alert system** instead of flat threshold alerts. Categorize waste into:

- **Critical**: Campaigns with 15+ clicks / 0 orders AND daily spend > $10 — these are actively bleeding money

- **Warning**: ACoS 40-60% but with some conversions — needs optimization, not killing

- **Watch**: Campaigns trending toward waste (ACoS climbing week-over-week)

This prevents alert fatigue for your PPC manager. They’ll know exactly which fires need immediate action vs. what can wait for the weekly deep-dive. The Supabase historical data makes the “trending toward waste” detection possible from Week 1 itself.

-–

### Question

For the weekly royalty CSV upload in Phase 3 — is this a manual process where someone drops the file, or does it land somewhere automatically (Google Drive folder, email attachment, S3 bucket)? This determines whether I build a webhook trigger with a simple upload form, a Google Drive file-watch node, or an email parser. It also affects whether the Monday morning brief can be fully automated or needs a human trigger.

-–

Happy to do a quick 15-minute call to walk through my technical approach in detail.

**Priyanshu Kumar**

AI Consultant — n8n + Cl

aude AI Specialist

[email protected]

17

Hi Colin,
About me: I’m a developer based in Europe with hands-on experience building n8n production workflows that integrate REST APIs, AI models, and messaging platforms. I’ve worked with OAuth 2.0 flows, API pagination/rate limiting, and structured LLM output parsing in n8n — exactly the skill set this project needs.

Relevant experience:

  • Built n8n workflows integrating multiple APIs (REST/OAuth 2.0) with AI processing (Claude & GPT) and structured Slack/messaging output

  • Experience with Supabase/PostgreSQL for metric storage and week-over-week comparisons

  • Designed Claude prompt chains for business data analysis with structured JSON output parsing

Timeline: 3 weeks as specified Fixed price: $2,200 USD (phased 33/33/34% as outlined)

My question about the brief: For the Pricing Optimizer — you mention flagging designs at $13.38 that have BSR traction. What BSR threshold defines “traction” for your catalog? (e.g., under 500K? under 1M?) This will be critical for tuning the Keepa query filters and avoiding false positives that flood the #listings-tasks channel with noise.

Looking forward to discussing further.Best, Marco

18

Hi Colin,

Re: Business Performance Analyzer

This is a well-scoped brief — the phased structure, clear data sources, and defined Slack outputs make it straightforward to execute confidently. Here’s my relevant background:

On the LLM side, I work with Claude API regularly — building structured prompt pipelines, parsing JSON outputs, and using Haiku/Sonnet strategically based on task complexity, exactly the pattern you’ve described for daily scans vs weekly analysis. I’ve built multi-agent systems where different models handle different reasoning layers in production.

On the data pipeline side, I’ve engineered real-time streaming architectures processing 100k+ events/second using GCP Pub/Sub, Dataflow, and BigQuery — so designing Supabase tables for week-over-week metric storage and building reliable scheduled workflows is well within my range.

On the API integration side, I’ve worked extensively with OAuth 2.0, REST APIs, webhook handling, pagination, and rate limiting across healthcare, automotive, and SaaS platforms. I’ve integrated WhatsApp Business API, Twilio, and multiple third-party services into production systems.

I haven’t worked with Keepa or Amazon Ads API specifically, but both are well-documented REST APIs — the Amazon Ads API OAuth flow is standard, and Keepa is straightforward. I’d have both integrated within the first two days of Week 1.

My one question/suggestion: For the Pricing Optimizer, are BSR snapshots from Keepa being pulled per ASIN individually, or are you batching by niche/tag from ClickUp? Batching by niche would significantly reduce Keepa API calls and keep you well within the €19/mo plan limits — worth confirming before I finalize the workflow architecture.

Timeline and fixed price: Phase 1 (Week 1) — $850 Phase 2 (Week 2) — $850 Phase 3 (Week 3) — $900 Total: $2,600 fixed

I’m available to start immediately. Happy to jump on a discovery call this week.

Ishola — NeuralicAI [email protected] GitHub: Neuralic (Neuralic) · GitHub

my website: neuralic-ai.vercel.app