Sample savings report

Your AI bill has $4,812/month in testable savings.

SpendLens AI turns raw LLM calls into a clear action list: which endpoint to test, which model to try, how much it saves, and how confident the recommendation is.

Biggest move this week

Replay product recommendations on Gemini Flash.

Estimated savings: $2,140/month. Confidence: High. Start with a replay, then roll out to 10% of traffic.

Projected monthly spend

$12,654

Based on 18,291 calls yesterday

Recommended savings

$4,812/mo

38% lower run rate if accepted

High-confidence savings

$3,284/mo

Ready for replay testing this week

Payback signal

1 day

Pro plan covered by first accepted switch

Recommendation queue

Savings by endpoint

recommend-products

High

shopping/catalog.py::recommend_products

Current: gpt-4o
Recommended: Gemini Flash
Spend: $5,210
Savings: $2,140/mo

Compact product-ranking prompts and stable catalog fields make this a strong replay candidate.

summarize-product-reviews

High

shopping/reviews.py::summarize_reviews

Current: claude-sonnet-4
Recommended: Qwen Turbo
Spend: $3,420
Savings: $1,144/mo

Summaries are under 500 tokens and quality risk is low after replay.

extract-shopping-filters

Medium

shopping/search.py::extract_filters

Current: gpt-4o
Recommended: DeepSeek Chat
Spend: $1,980
Savings: $792/mo

Structured filter extraction looks safe, but needs validation on ambiguous shopping queries.

compare-products

Low

shopping/compare.py::compare_products

Current: claude-sonnet-4
Recommended: Llama 3.1 70B
Spend: $2,044
Savings: $736/mo

Open-ended answers need human review before any rollout.

Endpoint	Current	Recommended AI	Spend	Savings	Confidence
recommend-products shopping/catalog.py::recommend_products	gpt-4o	Gemini Flash	$5,210	$2,140/mo	High Compact product-ranking prompts and stable catalog fields make this a strong replay candidate.
summarize-product-reviews shopping/reviews.py::summarize_reviews	claude-sonnet-4	Qwen Turbo	$3,420	$1,144/mo	High Summaries are under 500 tokens and quality risk is low after replay.
extract-shopping-filters shopping/search.py::extract_filters	gpt-4o	DeepSeek Chat	$1,980	$792/mo	Medium Structured filter extraction looks safe, but needs validation on ambiguous shopping queries.
compare-products shopping/compare.py::compare_products	claude-sonnet-4	Llama 3.1 70B	$2,044	$736/mo	Low Open-ended answers need human review before any rollout.

Savings graph

Monthly savings by workload

Product recommendations$2,140

Review summaries$1,144

Shopping filters$792

Product comparisons$736

$4,812/month total

The first two high-confidence changes cover most of the upside without asking the team to touch risky customer-facing QA yet.

Prompt Waste Signals

Token leaks that do not require a model switch.

These are advisory, rule-based signals from existing metadata. SpendLens does not rewrite prompts automatically or send full user prompts to external services.

Large prompt/context detected

High

Workload: rag-answer

Avg input

18,400

Avg output

640

Review whether all retrieved chunks are needed. Try reducing RAG chunks, filtering context more aggressively, or using prompt caching if available.

Long output detected

Medium

Workload: write-product-description

Avg input

4,900

Avg output

2,300

Add concise-output instructions or max token limits where appropriate.

Large repeated prompt template

Medium

Workload: recommend-products

Avg input

3,200

Avg output

Shorten static instructions or move reusable context outside the repeated prompt.

Confidence score

Know what to trust first.

High

Move to replay testing now.

Medium

Test carefully on edge cases.

Low

Keep as manual review only.

Suggested next actions

A savings report your engineers can actually ship.

01Replay 200 recent examples on the high-confidence model-switch recommendation.
02Reduce RAG chunks on rag-answer and compare answer quality against current output.
03Add concise-output instructions to write-product-description and cap max output tokens.
04Ship the high-confidence model switch to 10% of traffic.
05Watch cost, latency, disagreement, and prompt-token averages for 24 hours.