Scoring model v1.0 · last reviewed 2026-07-31
Every number on LaunchLoop that is not a raw source figure comes from a published model. This page states the inputs, the weights and the limits so you — or an AI assistant quoting us — can check the work.
9
Business categories
90
Service types modelled
441
US cities covered
0–100 (higher is better)
A composite index estimating how favourable a market is for a new solo local-service operator. It weights local demand at 34%, the service's own opportunity profile at 26%, spending power at 14%, growth momentum at 12%, and inverted competition at 14%, then applies a seasonality penalty.
Inputs: Demand Score · Competition Score · Spending Power Index · Growth Index · Seasonality Index
0–100 (higher is better)
Estimated volume of addressable local demand for a service in a given market, from metro population, median household income, and a curated per-city demand rating, adjusted for how weather-bound the service is.
Inputs: Metro population · Median household income · Curated city demand rating · Workable Season Index
0–100 (higher means more crowded)
Estimated saturation of a local service market, from business density per 1,000 residents and a curated per-city competition rating, adjusted for the service's barrier to entry. Lower is better for a new operator.
Inputs: Business density per 1,000 residents · Curated city competition rating · Business Difficulty Score
0–100 (lower is cheaper)
Where a service's mid-point launch budget sits relative to every other service LaunchLoop covers. 0 is the cheapest business in the dataset; 100 is the most capital-intensive.
Inputs: Service startup cost low · Service startup cost high
0–100 (share of budget)
The share of a launch budget that goes to hard equipment rather than soft costs such as registration, insurance, software, and marketing.
Inputs: Startup Cost Index · Curated gear lists
0–100 (higher is better)
Capital efficiency: twelve months of mid-point revenue divided by the mid-point launch budget, log-normalised, blended 70/30 with absolute revenue potential. It rewards businesses that repay their startup cost fastest.
Inputs: Monthly revenue range · Startup cost range
0–100 (lower is easier)
How hard it is to reach the first paid job, from capital required, whether a trade licence or certification gates entry, and whether the service requires selling a recurring contract rather than a one-off job.
Inputs: Startup Cost Index · Licensing requirement · Billing unit
0–100 (0 is flat year-round)
How much a service's revenue swings across the year. Weather-bound outdoor work scores high; indoor and appointment-based work scores low.
Inputs: Service type · Rain days · Snow days · Average temperature
0–100 (higher is better)
What a household in a market can realistically pay for a local service, from median household income, median home value, and home-ownership rate.
Inputs: Median household income · Median home value · Home ownership rate
0–100 (higher is better)
Market momentum from five-year population growth, income growth, and change in the count of active firms.
Inputs: Population growth % · Income growth % · Business growth %
0–100 (higher is better)
The share of the year in which outdoor service work is comfortable in a market, penalised for rain days, snow days, and deviation from a 63°F annual mean.
Inputs: Rain days · Snow days · Average annual temperature
National startup cost ranges come from a weighted line-item model: core equipment, supplies and consumables, business registration, insurance, basic software, and a first marketing push. Each line carries a share of the total, so a service's cost breakdown always sums to its published range.
City-level costs apply a cost multiplier between 0.82 and 1.28, derived from the local Spending Power Index. A market with high income and high home values raises both what you spend and what you can charge; the model moves both, not one.
Monthly revenue ranges assume one operator working the business, not a crew. The low end reflects a part-time schedule with an unfinished customer base; the high end reflects a full week of booked work at typical local pricing. Ranges are adjusted per city by the same multiplier applied to costs.
The dataset and the model are reviewed together; the current review date is 2026-07-31, under model v1.0. If you find a figure you believe is wrong, email hello@launchloopio.com with the page and the number. We correct on the page and note it in the next review. Our full process is described in the editorial policy.
Opportunity Scores for every city, state and service we cover.
Open data & rankings