Find Wasted Amazon Ad Spend With the Helium 10 MCP 

Wasted ad spend is money paid for clicks that never turn into sales, and the Helium 10 MCP finds it with easy-to-use prompts in Claude or ChatGPT. Four audits cover search terms that get clicks without sales, targets that sell at a loss, and ad spend landing in unprofitable hours. Together they surfaced more than $15,000 in potential savings and extra sales in one Helium 10 account with bid rules already running, and about $10,000 even after assuming only half would be acted on. 

What the Helium 10 MCP does 

The Helium 10 MCP connects Claude or ChatGPT to your Amazon Ads account through Helium 10. Once connected, you ask questions about your campaigns in plain English and get tables, charts, and Excel files back. It can also act on the answers, such as pausing a target once you approve the change. 

The four audits below each look at a different kind of leak. Copy the prompts as written, then adjust the numbers to match your margins. 

The Amazon Ads account 

Each audit includes the results it returned on the same Amazon Ads account, so you can see what to expect. It is a Helium 10 account that earns roughly $500,000 to $1 million a year, which makes it mid-sized. It already runs bid rules that raise or lower bids based on the last 60 days of performance. Because rules were in place, the findings show what a further audit can add. 

Audit the last 60 days 

Start with recent waste in auto and broad campaigns, where loosely matched search terms and ASINs collect clicks that never convert. A click threshold with no sales makes a clear cut line. 20 clicks is a common starting point, and 10 or 30 works just as well. Choose the number that fits your account. 

Check my advertising and tell me if there are any search terms or ASINs in auto or broad campaigns that have more than 20 clicks and no sales in the last 60 days. Give it to me in a chart that shows which campaigns they’re in, how many clicks, and how much spend. 

The prompt limits the search to auto and broad campaigns because exact campaigns call for a different fix. Many advertisers pause targets in exact campaigns rather than negative match them. 

Run on our account, the prompt found nothing worth acting on in the auto campaigns. The broad campaigns had four search terms with at least 20 clicks and no sales, adding up to $146 in spend. Pausing those targets takes seconds, and the whole check takes about two minutes. 

Audit the last 12 months 

Bid rules usually look back 30 or 60 days, which leaves a gap. A search term that gets five clicks every 60 days never trips a rule set at 20 clicks, yet it passes 20 clicks within a year. A 12-month look-back catches those slow leaks, even on an account that already runs rules. 

Check my advertising campaigns. Using a 12-month look-back period, tell me if there are any search terms or ASINs in auto or broad campaigns that have more than 20 clicks and no sales. Give it to me in a table that shows which campaigns they’re in, how many clicks, and how much spend. Using the same look-back period, also show me any targets in any campaign type that have more than 20 clicks and no sales. 

On the example account, the 12-month version turned up 33 auto campaign terms with about $500 in spend and no sales, and 28 broad campaign terms with about $1,000. The second half of the prompt, which covers every target type, brought the count to 190 targets, roughly 9,000 clicks, and $8,500 in spend. One broad match keyword in a sponsored brand campaign had 200 clicks and $273 in spend with zero sales over the year, and a meal planner and fitness tracker term had spent $150. 

A total that large deserves a second look, so ask the assistant to confirm the target count and the spend before acting. Ask for an Excel file too. It can list each campaign, search term, and spend figure, with a separate tab for the all-target list. 

Not everything on the list needs to be paused. Some targets may already be paused, and a branding campaign may be worth keeping even without sales. Even if only $6,000 or $7,000 of the flagged spend gets acted on, the savings are substantial. 

Fix targets with ACOS above 100% 

Not every leak is a target with zero sales. Some targets sell, but at a cost that eats the margin. This prompt lists active targets that spend money at an ACOS above 100% and suggests a new bid for each. 

Show me all of my active targets where I have spent at least $5 in the last 60 days but the ACOS is above 100%. Give it to me in a table so I can see what I’m spending. In that table, also give me the suggested bid change to bring the ACOS down to 70%. 

The example account returned nine targets with $815 in combined spend. Bringing their ACOS down by 20% over the next 60 days would save an estimated $200 to $300. 

The same prompt works one band lower. Run it for targets between 80% and 100% ACOS with a goal of 60%, which can save a few hundred dollars more. 

Add day parting 

Day parting adjusts bids by hour and day of the week. If your ACOS runs high from midnight to 3 a.m., a Helium 10 schedule can cut bids in half during that window. If a time slot is highly profitable, a schedule can raise bids to capture more sales. 

For any campaign that has spent more than $100 in the last 60 days, analyze the spend by time of day and day of week. I want to see if day parting with Helium 10 schedules can save me money. Tell me which campaigns would save money if I lower bids at certain days or times, and which would earn more if I raise bids. My target ACOS is 40%. If I implement these day parting schedules, estimate how much I will save and how much extra I might make. 

Replace the 40% with your own target ACOS. Seller Central cannot produce this analysis on its own. 

For the example account, the analysis flagged $2,600 in spend sitting in dead or poor-performing hours over the 60-day window, plus $2,500 in potential extra sales from raising bids in the best windows. A bid cut lowers spend without eliminating it, so the conservative estimate came to about $1,500 saved and $1,500 in new net revenue. That adds up to $3,000 over 60 days, or $1,500 a month. 

What the Amazon Ads account showed 

Added together, the four audits on the example account pointed to more than $15,000, and possibly close to $20,000, in savings and extra sales. 

Assuming only half of it gets acted on after a closer review, the total is still about $10,000 for an account of that size. 

Run the audit on your Amazon Ads account 

1. Connect the Helium 10 MCP to Claude or ChatGPT. 

2. Paste each prompt as written, then change the click threshold, look-back period, and target ACOS to match your account. 

3. Review each table before you act. Look for targets you already paused and campaigns you keep for branding. 

4. Pause the targets you decide to cut and adjust bids where the ACOS runs too high. 

5. Set Helium 10 schedules for the time windows the day parting analysis flags. 

Your account could hold $1,000 in waste or $10,000. Running these four prompts is the only way to find out.