# Reaktion MCP prompt library

44 tested prompts for the Reaktion MCP, grouped by what you want to decide. Updated 2026-11-02. Latest version: https://mcp.reaktion.com/prompts.md

## How to use this file

1. Connect the Reaktion MCP in Claude, ChatGPT, Grok or Gemini. Server URL: `https://api.reaktion.com/mcp`. Setup guides: https://mcp.reaktion.com/#connect
2. Paste a prompt and name your store or client, for example "for Nordic Living".
3. Or save this file in a Claude Project or ChatGPT project. Your assistant can then pick the right prompt when you describe what you need.

The MCP is included in every Reaktion subscription at no extra cost. It is read-only: it never changes campaigns, budgets or flows.

## Contents

- Executive reporting (5)
- Profitability (5)
- Google Ads (6)
- Paid social and creative (5)
- Customers and retention (5)
- Products and inventory (5)
- Analytics and CRO (5)
- Klaviyo and messaging (4)
- Agencies (4)

## Executive reporting

### Daily ecommerce performance pulse

See yesterday's commercial performance and exceptions.

```text
Compare yesterday with the same weekday last week for every connected store. Show orders, sales excluding VAT after refunds, gross profit, net profit, and ad spend. Explain changes and flag incomplete recent data.
```

**Metrics:** Orders · Total Sales, excl tax, after refunds · Gross Profit, excl tax, after refunds · Net Profit, excl tax, after refunds · Ad Spend  
**Dimensions:** Shop  
**Needs connected:** Connected store and ad accounts; net profit needs cost configuration

### Weekly growth brief

Summarize what grew, fell or needs action.

```text
Give me a weekly ecommerce briefing, comparing the last seven completed days with the prior seven. Break down sales, gross profit, spend and conversions by shop and marketing account. Include 3 opportunities and 3 risks.
```

**Metrics:** Total Sales, excl tax, after refunds · Gross Profit, excl tax, after refunds · Ad Spend · ROAS, Primary, before refunds, incl tax  
**Dimensions:** Shop · Data Source Account  
**Needs connected:** Connected ecommerce and marketing integrations

### Which channels drive profitable growth?

Compare channel scale versus economic contribution.

```text
For the last 30 completed days, compare channel spend and primary attributed revenue, ROAS, and POAS. Also show store-level MER separately if supported. Explain that platform conversions can overlap across channels.
```

**Metrics:** Ad Spend · ROAS, Primary, before refunds, incl tax · POAS, Primary, before refunds, incl tax  
**Dimensions:** Data Source Type · Channel Attribution (Last Click)  
**Needs connected:** Marketing account(s) and conversion tracking; do not sum overlapping attribution sources

### Find unusual performance changes

Spot campaign and store changes that deserve attention.

```text
Compare the last three completed days with matching weekdays in the previous week. Flag unusually large changes in spend, orders, conversion rate, gross profit and refunds. Show evidence and caveats before recommendations.
```

**Metrics:** Ad Spend · Orders · Gross Profit, excl tax, after refunds · Store Conversion Rate · Return Rate  
**Dimensions:** Day · Data Source Account  
**Needs connected:** Enough historical data; compare equivalent periods

### Compare Denmark and Sweden

Compare market-level growth with consistent reporting currencies.

```text
Compare performance by connected shop and country for the last 30 completed days. Show orders, sales excluding VAT after refunds, gross profit and ad spend. Use the organisation reporting currency, label countries and note FX effects.
```

**Metrics:** Orders · Total Sales, excl tax, after refunds · Gross Profit, excl tax, after refunds · Ad Spend  
**Dimensions:** Shop · Country  
**Needs connected:** At least two shops or countries with comparable cost setups

## Profitability

### Explain where profit went

Connect revenue to variable cost, advertising and custom costs.

```text
Build a profit bridge for last month: sales after refunds excluding VAT, configured variable costs, gross profit, ad spend, custom cashflows and net profit. Identify which drivers explain margin changes from the prior month.
```

**Metrics:** Total Sales, excl tax, after refunds · Gross Profit, excl tax, after refunds · Net Profit, excl tax, after refunds · Ad Spend  
**Dimensions:** Month · Shop  
**Needs connected:** Complete cost and cashflow settings

### Find products losing money to refunds

Identify commercial impact of returns.

```text
Rank products by refunded units and refunded product sales over the last 90 completed days. Compare against gross profit and units sold where compatible. Identify high-return products and avoid treating refunds on older orders as new orders.
```

**Metrics:** Return Rate · Refunded Product Sales, excl tax · Units Sold, after refunds · Gross Profit, excl tax, after refunds  
**Dimensions:** Product · Product Variant  
**Needs connected:** Refund integration and chosen refund date method

### Are discounts destroying margin?

See how promotions affect profits.

```text
Compare the last 30 days by discount-code usage and store. Show sales, variable costs, gross profit and refunds. Highlight promotions that add volume but dilute profit, distinguishing discounts from refunds.
```

**Metrics:** Gross Profit, excl tax, after refunds · Total Sales, excl tax, after refunds · Return Rate  
**Dimensions:** Discount code · Shop  
**Needs connected:** Order discount-code coverage and costs

### Compare profit after marketing

See contribution after media cost.

```text
For the last completed month, report gross profit excluding VAT after refunds, total ad spend, and gross profit after ad spend. Show a prior-month comparison and separate store profit from attributed campaign profit.
```

**Metrics:** Gross Profit, excl tax, after refunds · Ad Spend · Net Profit, excl tax, after refunds  
**Dimensions:** Shop  
**Needs connected:** Marketing spends and cost settings

### Which variable costs increased?

Find whether COGS, payment or shipping drives changes.

```text
Investigate changes in total variable cost share between this and the previous completed month. Show shipping and payment processing costs separately, and explain possible reasons without assuming cause.
```

**Metrics:** Gross Profit, excl tax, after refunds · Total Sales, excl tax, after refunds  
**Dimensions:** Shop · Month  
**Needs connected:** Store cost configuration, expense inputs

## Google Ads

### Find profitable Google Ads scaling opportunities

Combine business outcomes with constrained search reach.

```text
Compare Google Search campaigns for the last 30 completed days by ad spend, Primary ROAS and POAS. Include Search Impression Share, Lost IS Budget and Lost IS Rank only when their reporting basis is validated. Identify campaigns with profitable results and credible headroom to scale; do not recommend budget changes from invalid or missing share data.
```

**Metrics:** Ad Spend · ROAS, Primary, before refunds, incl tax · POAS, Primary, before refunds, incl tax · Search Impression Share · Lost Search IS, Budget · Lost Search IS, Rank  
**Dimensions:** Campaign  
**Needs connected:** Google Ads Search data; QA hold on impression share sums

### Budget constraint or Ad Rank?

Distinguish visibility loss mechanisms.

```text
For each eligible Search campaign, analyze share lost to budget versus share lost to rank with spend, conversions and ROAS. Classify whether the next step is measurement QA, budget, creative/landing relevance or bidding review.
```

**Metrics:** Search Impression Share · Lost Search IS, Budget · Lost Search IS, Rank · Ad Spend · ROAS, Primary, before refunds, incl tax  
**Dimensions:** Campaign  
**Needs connected:** Google Search; validated non-additive share semantics

### Find expensive search terms without orders

Spot paid queries draining spend.

```text
Group the last 30 days of Google Ads results by search term. Rank terms with meaningful spend and no Primary purchase conversions. Suggest negatives only as candidates and clearly identify brand or strategic terms.
```

**Metrics:** Ad Spend · Clicks · Cost per Order Conversion  
**Dimensions:** Marketing Search Term  
**Needs connected:** Google Ads query data; zero conversions may reflect attribution delay

### Rank campaigns by profit efficiency

Prioritize spend decisions, not just ROAS.

```text
For the last 30 completed days, rank Google Ads campaigns by Primary POAS, spend, conversion volume and profit after ad spend. Separate low-data campaigns and explain the tradeoffs.
```

**Metrics:** Ad Spend · POAS, Primary, before refunds, incl tax · Cost per Order Conversion  
**Dimensions:** Campaign  
**Needs connected:** Primary conversions and matching spend data

### Brand versus nonbrand performance

Separate navigational demand from prospecting.

```text
Using campaign names and available branded status, compare brand and nonbrand Google Search campaigns by spend, conversions, ROAS and POAS. Explain any ambiguous classification.
```

**Metrics:** Ad Spend · ROAS, Primary, before refunds, incl tax · POAS, Primary, before refunds, incl tax  
**Dimensions:** Campaign  
**Needs connected:** Reliable campaign naming and/or campaign branded attribute

### Which campaigns changed yesterday?

Prioritize sudden spend/conversion shifts.

```text
Compare yesterday with the same weekday last week for each Google Ads campaign. Show spend, clicks, conversion volume, ROAS and POAS and highlight changes with enough volume to matter.
```

**Metrics:** Ad Spend · Clicks · Cost per Order Conversion · ROAS, Primary, before refunds, incl tax  
**Dimensions:** Campaign · Day  
**Needs connected:** Recent marketing data may continue to update

## Paid social and creative

### Find the creatives driving profitable sales

Combine creative view with economic performance.

```text
Analyze Meta ads in the last 45 days with a meaningful spend threshold. Show ad name, thumbnail when available, spend, Primary conversions, revenue, ROAS, POAS and CTR. Identify winners and explain the evidence.
```

**Metrics:** Ad Spend · Click-Through Rate · ROAS, Primary, before refunds, incl tax · POAS, Primary, before refunds, incl tax  
**Dimensions:** Ad · Image Creative  
**Needs connected:** Meta Ads, creative metadata, primary conversion config

### Detect potential ad fatigue

Spot weakening attention and increasing costs.

```text
Compare Meta creatives over the latest two consecutive 14-day periods. Evaluate spend, reach, frequency, CTR, cost per order and POAS. Flag fatigue candidates, but do not conclude fatigue from frequency alone.
```

**Metrics:** Average Frequency · Click-Through Rate · Cost per Order Conversion · POAS, Primary, before refunds, incl tax  
**Dimensions:** Ad · Image Creative  
**Needs connected:** Enough Meta delivery volume and attributed conversions

### Which videos retain attention?

Compare full-play depth and downstream results.

```text
Rank video creatives by impressions, 25% and 100% video completions, CTR, spend and attributed POAS. Present video retention proxy separately from sales efficiency and skip unavailable playback metrics.
```

**Metrics:** Video Views, 25% · Video Views, 100% · Click-Through Rate · POAS, Primary, before refunds, incl tax  
**Dimensions:** Video Creative · Ad  
**Needs connected:** Video assets and platform video-view metrics

### Which hooks create profitable attention?

Analyze creative hook annotations against commercial results.

```text
Group available image creatives by hook tactic or visual-format annotation. Compare exposure, clicks, attributed revenue and POAS. Report sample counts and avoid assigning causation to hook labels.
```

**Metrics:** Impressions · Clicks · POAS, Primary, before refunds, incl tax  
**Dimensions:** Image Creative · Ad  
**Needs connected:** Image annotation coverage and creative attributes

### Audit thumb-stop reporting

Validate an attention metric before using it.

```text
List the source formula, description and daily reported value for Thumb Stop Ratio. Compare it with available short-play metrics, and mark any 3-second versus 30-second mismatch as an unresolved definition issue.
```

**Metrics:** Thumb Stop Ratio · Impressions  
**Dimensions:** Ad · Video Creative  
**Needs connected:** Formula validation needed; do not interpret until fixed

## Customers and retention

### Which campaigns acquire profitable new customers?

Use new-customer profit, not just sales.

```text
Compare campaigns by new-customer attributed revenue, new-customer profit, New Customer POAS and spend over the last completed month. Keep conversion reporting source and refund basis explicit.
```

**Metrics:** New Customer Revenue, after refunds · New Customer Profit, after refunds · New Customer POAS, before refunds  
**Dimensions:** Campaign  
**Needs connected:** Primary conversion and new-customer identity mapping

### Which customer segments need attention?

Examine customer value through RFM groups.

```text
Break down customer activity and order value by RFM segment for the last 90 completed days. Identify growing high-value and declining segments without exposing personal identifiers.
```

**Metrics:** Orders · Total Sales, excl tax, after refunds  
**Dimensions:** Customer RFM Segment  
**Needs connected:** Customer identity stitching and RFM segmentation

### How dependent are we on returning customers?

Assess repeat business mix.

```text
Compare first-time versus repeat-customer performance where data supports the distinction. Show order count, profit and last-month change; explain denominator and classification method.
```

**Metrics:** Orders · Gross Profit, excl tax, after refunds  
**Dimensions:** Customer · Month  
**Needs connected:** Reliable customer identity and order history; avoid publishing the anomalous Returning Customer Order Rate formula

### Which acquisition channels create valuable customers?

Compare attributed customer lifetime value.

```text
By campaign or channel, compare initial new-customer results with Reaktion Web attributed customer lifetime value over the chosen lookback period. State the CLV definition and attribution coverage before ranking sources.
```

**Metrics:** Customer Lifetime Value (Web) · New Customer Revenue, after refunds  
**Dimensions:** Campaign · Channel Attribution (Last Click)  
**Needs connected:** Maturity of customer cohorts; CLV attribution windows

### What distinguishes high-value customers?

Explore purchase frequency and product preference.

```text
Compare customer segments using order history, product types and gross profit. Aggregate findings by segments and product groups, not customer names, unless the user is specifically authorized to view customer-level records.
```

**Metrics:** Orders · Gross Profit, excl tax, after refunds  
**Dimensions:** Customer RFM Segment · Product Type  
**Needs connected:** Customer segmentation; adequate group sizes

## Products and inventory

### Find the products that actually drive profit

Rank by profit, not just sales.

```text
Rank products by sales excluding VAT, units sold, gross profit and refunds over the last 90 days. Show both product-level and variant-level analysis if supported. Distinguish contribution before ad spend.
```

**Metrics:** Product Sales, excl tax, before refunds · Units Sold, after refunds · Gross Profit, excl tax, after refunds · Return Rate  
**Dimensions:** Product · Product Variant  
**Needs connected:** Product cost coverage; appropriate product attribution

### Find slow-moving stock

Prioritize unsold inventory that ties up capital.

```text
Find product variants with inventory but weak recent sales. Show SKU, quantity, retail value, sell-through rate and potential gross profit exposure. Do not treat current inventory values as historical inventory.
```

**Metrics:** Sell-Through Rate · Inventory Retail Value · Units Sold, after refunds  
**Dimensions:** Product Variant  
**Needs connected:** Shopify stock levels and available variant cost/price attributes

### Find SKUs with disproportionate returns

Identify size, fit or quality risks.

```text
Rank SKUs by refunded units divided by sold units over the last 90 days. Require a minimum sales sample, show absolute unit counts and highlight products with high refund value.
```

**Metrics:** Return Rate · Refunded Product Sales, excl tax · Units Sold, after refunds  
**Dimensions:** Product Variant SKU · Product Variant  
**Needs connected:** Refunded quantity coverage; avoid small-sample rankings

### Which products win across stores?

Compare matched products across markets.

```text
Compare cross-store matched products by units, sales, gross profit and return rate, broken down by shop and country. Note any product matching limitations.
```

**Metrics:** Units Sold, after refunds · Gross Profit, excl tax, after refunds · Return Rate  
**Dimensions:** Product (Cross-Store) · Shop  
**Needs connected:** Cross-store product mapping and currency normalization

### Are sale products really more profitable?

Analyze promotion economics at variant level.

```text
Compare sale versus non-sale product variants by sales, units, gross profit and returns. Avoid assuming discounted orders are incremental demand.
```

**Metrics:** Product Sales, excl tax, before refunds · Gross Profit, excl tax, after refunds · Return Rate  
**Dimensions:** On-sale Status · Product Variant  
**Needs connected:** Sale-status accuracy; discount and order data

## Analytics and CRO

### Which landing pages receive traffic but fail to sell?

Locate traffic-to-order leakage.

```text
For the last 30 completed days, rank landing pages by sessions and available conversion measurements. Identify high-traffic pages with weak conversion. Reconcile GA4 purchase data with Shopify orders before treating zero revenue as zero sales.
```

**Metrics:** Sessions · Store Conversion Rate · Total Sales per Session  
**Dimensions:** Analytics Landing Page  
**Needs connected:** GA4 sessions, landing page dimension; cross-domain metric may not support same grain

### Which traffic sources attract valuable visitors?

Compare engagement and order outcomes.

```text
Group analytics by source and channel group. Show sessions, engaged sessions, bounce rate and purchase metrics where genuinely available. Explain discrepancies between traffic data and store sales.
```

**Metrics:** Sessions · Bounce Rate · Store Conversion Rate  
**Dimensions:** Analytics Source · Analytics Channel Group  
**Needs connected:** GA4 source/channel dimensions and tracking

### Does mobile traffic underperform?

Investigate device conversion and quality.

```text
Compare sessions, bounce rate and supported purchase metrics by analytics device type. Break down countries when sample sizes allow. Provide testable hypotheses, not guaranteed fixes.
```

**Metrics:** Sessions · Bounce Rate · Store Conversion Rate  
**Dimensions:** Analytics Device Type · Country  
**Needs connected:** GA4 device data, compatible conversion metric grain

### Which markets monetize traffic best?

Compare shop commerce relative to traffic.

```text
For the last completed month, show sales per session by shop and country, alongside session volume and store conversion rate. State that sessions come from GA4 while sales/orders come from the ecommerce platform.
```

**Metrics:** Total Sales per Session · Sessions · Store Conversion Rate  
**Dimensions:** Shop · Country  
**Needs connected:** Same shop coverage in GA4 and ecommerce

### Are tracking signals missing?

Investigate disagreements across measurement sources.

```text
Compare GA4 orders, ecommerce orders and marketing conversion data over the same dates. Identify possible gaps and explain that different attribution methods, consent and event delays can create legitimate differences.
```

**Metrics:** Orders · Store Conversion Rate · Tracking Uplift  
**Dimensions:** Day · Shop  
**Needs connected:** Aligned measurement windows and event definitions

## Klaviyo and messaging

### Which email flows drive attributed revenue?

Compare email conversion value and engagement.

```text
Rank Klaviyo flows and sendouts over the last 30 completed days by delivered messages, unique opens, unique clicks and Primary attributed revenue where available. Keep revenue basis and attribution source explicit.
```

**Metrics:** Email Deliveries · Email Open Rate · Email Click-to-Open Rate  
**Dimensions:** Email/SMS Flow · Email/SMS Sendout  
**Needs connected:** Klaviyo connected; messaging conversion data

### Are email campaigns reaching customers?

Diagnose bounce and unsubscribe signals.

```text
Compare Klaviyo sendouts by deliveries, bounce rate, open rate and unsubscribe rate. Exclude zero-denominator rates and flag when opens exist but no same-period deliveries.
```

**Metrics:** Email Deliveries · Email Bounce Rate · Email Open Rate · Email Unsubscribe Rate  
**Dimensions:** Email/SMS Sendout  
**Needs connected:** Sendout and message delivery events

### Which automations could improve?

Spot flows with weak engagement or conversion.

```text
Compare supported Klaviyo flows by engagement and revenue, grouped by flow. Identify flows to review, distinguishing low traffic from low performance. Do not assume AI can modify flows through a read-only reporting connector.
```

**Metrics:** Email Deliveries · Email Click-to-Open Rate  
**Dimensions:** Email/SMS Flow  
**Needs connected:** Flow metadata and deliveries

### Does SMS improve retention?

Review messaging results by channel.

```text
Compare SMS deliveries and clicks with email deliveries and clicks over the last completed month. Describe platform differences and avoid treating email and SMS attribution as mutually exclusive.
```

**Metrics:** SMS Deliveries · Email Deliveries · Email Open Rate  
**Dimensions:** Data Source Type · Email/SMS Sendout  
**Needs connected:** SMS event integration

## Agencies

### Which clients need my attention?

Review multi-client performance efficiently.

```text
Across my accessible client organisations, compare the last seven completed days with the previous seven. Rank clients by material declines in profit, order volume, or marketing efficiency. Show exactly which accounts need investigation; never merge client data.
```

**Metrics:** Orders · Gross Profit, excl tax, after refunds · Ad Spend · ROAS, Primary, before refunds, incl tax  
**Dimensions:** Data Source Account · Shop  
**Needs connected:** Agency access to multiple organisations

### Prepare a client performance review

Create a concise, decision-ready briefing.

```text
Prepare a meeting brief for the selected ecommerce client: last month versus previous month, sales and profit excluding VAT after refunds, marketing efficiency, new-customer contribution, risks, and five evidence-based talking points.
```

**Metrics:** Net Profit, excl tax, after refunds · ROAS, Primary, before refunds, incl tax · New Customer Profit, after refunds  
**Dimensions:** Data Source Account · Shop  
**Needs connected:** Client context, connected sources

### Audit data differences across stores

Surface measurement configuration issues.

```text
For the selected organisation, list connected data source accounts, their reporting currencies and recent account data availability. Identify missing integrations, suspicious zeros and differences that could stem from timezones or refund methods.
```

**Metrics:** Orders · Sessions · Ad Spend  
**Dimensions:** Data Source Account · Data Source  
**Needs connected:** Access to organisation account inventory and reporting

### Generate a consistent weekly client report

Use one evidence standard across clients.

```text
Create a seven-day executive summary with store sales excluding VAT after refunds, gross profit, ad spend, Primary POAS, top opportunities and caveats. Label any unavailable metric rather than substituting another definition.
```

**Metrics:** Total Sales, excl tax, after refunds · Gross Profit, excl tax, after refunds · Ad Spend · POAS, Primary, before refunds, incl tax  
**Dimensions:** Shop · Campaign  
**Needs connected:** Connected ecommerce and marketing integrations

## Reporting rules the MCP follows

1. Store financials (sales, gross profit, net profit) are reported excluding VAT and after refunds unless you ask otherwise. Say it in your prompt to be explicit.
2. Paid-channel ROAS and POAS come from attributed conversions, which can include VAT and are before refunds. They are labelled so you do not compare them with store profit by mistake.
3. Never add or average ratios. Percentages, shares, ROAS and POAS are recalculated from their parts.
4. Be explicit about time. Compare completed periods and matching weekdays, and expect the last few days to keep updating as refunds land.
5. Platforms overlap. Google and Meta can both claim the same order, so use store-level metrics such as Business-wise POAS when you need one total across channels.

Full metric, dimension and attribute catalog: https://mcp.reaktion.com