What you are looking at: Revenue flows in from the left — each band is a service line, sized by its share of trailing 12-month sales. Everything converges into Total Revenue at the centre, then fans back out to the right showing where that revenue goes: direct costs (labor, materials, vehicles), operating expenses (selling, G&A), and what remains as Net Income. The wider the band, the higher the dollar amount. Use this view to quickly see which cost categories are consuming the most of your revenue and what margin is left over after all expenses.
Revenue & Expense Forecast Report
AI-Powered Business Intelligence with Prediction Intervals
ImportantPortfolio Demo — Illustrative Data Only
This report is a demonstration of the AI forecasting system built by AI Business Science. The underlying data structure comes from a real pest control company; their data was used with explicit client permission for portfolio purposes. To protect client confidentiality, all dollar figures have been scaled by a constant factor and statistical noise has been injected — no individual data point reflects actual company revenue or expenses. Relative patterns (seasonality, year-over-year trends, forecast interval behaviour) are representative of real model outputs, but the absolute numbers are not.
Note: The following accounts had a negative or zero net total over the trailing 12 months and are excluded from the chart: Contra / Adjustments.
| Revenue by Service Line | ||
| Trailing 12 months — QuickBooks actuals | ||
| Service Line | 12-Month Total | % of Total |
|---|---|---|
| Residential Pest Control | $636,606 | 59.9% |
| Protected Environment | $150,747 | 14.2% |
| Termite / Wood Destroying | $136,352 | 12.8% |
| Commercial Pest Control | $78,177 | 7.4% |
| Mosquito Services | $35,917 | 3.4% |
| Scorpion Specialty Services | $12,250 | 1.2% |
| Other Services | $6,039 | 0.6% |
| Rodent Services | $4,621 | 0.4% |
| Bed Bug Services | $1,395 | 0.1% |
How to read the interval: The point forecast is the model’s single best estimate. The Stretch Goal is the upper bound of the 80% prediction interval (
hi_80) — the high end of what history says is likely. The Defensive Floor is the lower bound (lo_80) — the low end of what history says is likely. Actual revenue lands between these two bounds roughly 8 times out of 10; anything outside them occurs about 2 times out of 10 (10% above, 10% below).
| Key Revenue Metrics | |
| Pest Control Services — March 2026 | |
| Metric | Value |
|---|---|
| Latest Actual (Jan 2026) | $88,843 |
| AI Point Forecast (March 2026) | $83,692 |
| Stretch Goal — hi_80 (March 2026) | $98,173 |
| Defensive Floor — lo_80 (March 2026) | $69,212 |
| Implied YoY Growth vs Mar 2025 (if forecast holds) | +4.7% |
Revenue Forecast with Prediction Interval
Previous Forecast Performance
How accurate was our last forecast? Here’s how the AI model performed for January 2026:
| January 2026 Results | |
| Forecast vs Actual | |
| Metric | Value |
|---|---|
| Actual Revenue | $88,843 |
| AI Forecast (XGBoost) | $107,523 |
| Difference | $18,680 over |
| Error Rate | 21.0% |
Expense interval: The point forecast is the model’s best estimate of operating costs. The upper bound (
hi_80) is the high end of what history says is likely — a month tracking here warrants cost review. The lower bound (lo_80) is the low end — a month tracking here suggests favorable cost control. Costs fall between these bounds roughly 8 times out of 10.
| Key Expense Metrics | |
| Pest Control Services — March 2026 | |
| Metric | Value |
|---|---|
| Latest Actual (Jan 2026) | $29,567 |
| AI Point Forecast (March 2026) | $32,363 |
| Stretch High — hi_80 (March 2026) | $36,436 |
| Stretch Low — lo_80 (March 2026) | $28,290 |
| Implied YoY Change vs Mar 2025 (if forecast holds) | -4.6% |
Expense Forecast with Prediction Interval
Weather Outlook
General Weather Outlook for March 2026
| Weekly Forecast Summary | ||
| For operational planning | ||
| Week | Rain Outlook | Temperature |
|---|---|---|
| Mar 01 – Mar 07 | Mostly dry | Warm |
| Mar 08 – Mar 14 | Mostly dry | Mild |
| Mar 15 – Mar 21 | Mostly dry | Mild |
| Mar 22 – Mar 28 | Mostly dry | Mild |
| Mar 29 – Mar 31 | Mostly dry | Warm |
| Note: Weather forecasts may shift by several days. Use as general guidance only. | ||
Monthly Overview:
- No significant rain expected, 5 week(s) likely dry
- No extreme heat expected
Actual timing may shift by several days. Monitor daily forecasts for scheduling.
Weather Forecast
Weather Summary
| Expected Conditions | ||
| March 2026 | ||
| Metric | Value | Impact |
|---|---|---|
| Avg Temperature | 69F | Higher temps increase pest activity |
| Max Temperature | 94F | Extreme heat drives indoor pest pressure |
| Total Precipitation | 0.16 in | Affects mosquito breeding |
| Avg Humidity | 31% | High humidity favors many pest species |
| Avg Wind Speed | 12.5 mph | Affects flying pest behavior |
Daily Weather Detail
| Daily Forecast | ||||
| March 2026 | ||||
| Date | Temp | High | Precip | Humid |
|---|---|---|---|---|
| Mar 01 | 78F | 94F | 0.0" | 25% |
| Mar 02 | 76F | 90F | 0.0" | 22% |
| Mar 03 | 71F | 84F | 0.0" | 22% |
| Mar 04 | 71F | 84F | 0.0" | 22% |
| Mar 05 | 70F | 81F | 0.0" | 22% |
| Mar 06 | 67F | 77F | 0.0" | 26% |
| Mar 07 | 63F | 78F | 0.0" | 38% |
| Mar 08 | 61F | 73F | 0.0" | 56% |
| Mar 09 | 60F | 66F | 0.0" | 60% |
| Mar 10 | 61F | 67F | 0.0" | 55% |
| Mar 11 | 67F | 74F | 0.0" | 40% |
| Mar 12 | 70F | 77F | 0.0" | 34% |
| Mar 13 | 68F | 75F | 0.02" | 33% |
| Mar 14 | 68F | 80F | 0.0" | 30% |
| Mar 15 | 68F | 81F | 0.0" | 28% |
| Mar 16 | 68F | 80F | 0.02" | 30% |
| Mar 17 | 68F | 80F | 0.01" | 32% |
| Mar 18 | 67F | 79F | 0.01" | 32% |
| Mar 19 | 68F | 79F | 0.01" | 31% |
| Mar 20 | 69F | 81F | 0.0" | 27% |
| Mar 21 | 69F | 81F | 0.0" | 29% |
| Mar 22 | 69F | 80F | 0.04" | 32% |
| Mar 23 | 68F | 78F | 0.03" | 31% |
| Mar 24 | 69F | 80F | 0.01" | 29% |
| Mar 25 | 70F | 82F | 0.0" | 26% |
| Mar 26 | 70F | 81F | 0.0" | 26% |
| Mar 27 | 69F | 80F | 0.0" | 27% |
| Mar 28 | 69F | 80F | 0.0" | 26% |
| Mar 29 | 70F | 82F | 0.0" | 25% |
| Mar 30 | 71F | 83F | 0.0" | 24% |
| Mar 31 | 71F | 83F | 0.01" | 24% |
How Our AI Model Works
Our forecasting system uses an ensemble of machine learning models, each specifically trained on your historical revenue and expense data:
Historical Pattern Analysis: The model learns from years of your revenue and expense data to identify seasonal trends and business cycles
Weather Integration: Temperature, precipitation, and humidity forecasts are incorporated because they directly impact pest activity and service demand
Adaptive Learning: Models are trained specifically on your historical performance, making predictions highly relevant to your operations
Model Selection: Multiple modeling techniques (XGBoost, LightGBM, NBEATSx neural network) are evaluated each month via cross-validation, and the best-performing model is selected
Understanding the Prediction Interval
The 80% prediction interval shown in this report is a conformal prediction interval — a statistically rigorous bound with an empirical coverage guarantee.
What does 80% mean in practice?
| Bound | Interpretation | When to Use |
|---|---|---|
| Stretch Goal (hi_80) | There is roughly a 10% chance actual revenue exceeds this value | Target for a strong month with favorable conditions |
| Point Forecast | The model’s best single estimate | Primary planning number |
| Defensive Floor (lo_80) | There is roughly a 10% chance actual revenue falls below this value | Planning baseline if early signals suggest a soft month |
The interval is calibrated using rolling-window conformal prediction: the model’s forecasting errors across multiple historical holdout windows are used to set the bounds. Unlike standard error bars, these intervals make no assumption about the shape of the error distribution, which is important because monthly revenue can have one-off spikes or dips that violate normality.
Model Validation & Confidence
Cross-validation: Each model is evaluated over 4 rolling holdout windows before the final forecast is generated. The model with the lowest average MAPE across windows is selected
Weather Impact Modeling: Seasonal demand drivers like cooling degree days and extreme heat events are captured as model features
Continuous Improvement: Each month’s actual results are compared against prior forecasts in the “Previous Forecast Performance” section
Transparent Reporting: The “Previous Forecast Performance” section shows exactly how well predictions matched reality, including whether actuals fell within the 80% interval
Weather Features Used in Forecasting
| Feature | Description | Impact on Pest Activity |
|---|---|---|
| Cooling Degree Days (CDD) | Cumulative heat above 65F | Higher values increase pest activity |
| Heating Degree Days (HDD) | Cumulative cold below 65F | Drives pests indoors |
| Days Over 100F | Extreme heat count | Peak pest pressure periods |
| Total Precipitation | Monthly rainfall | Affects mosquito breeding |
| Average Humidity | Moisture levels | Favors many pest species |
Report generated on March 29, 2026 using AI forecasting technology by AI Business Science.