What Baseline Should You Save Before a New Marketing Provider Starts?

A new provider can change more than the marketing. It may change the pages, campaign focus, inquiry fields, and definitions used in reporting. Without a starting record, you can end up comparing two different measurements and calling the difference a business result.

Prepare a baseline before the first changes go live. It does not need to prove cause and effect. Its job is to preserve what the business was doing, what the numbers meant, and which conditions made that period unusual. That gives both parties a fairer reference for later decisions.

Save the definition beside every number

• Baseline item: Inquiry; Definition to preserve: Which requests count, and whether duplicates are removed

• Baseline item: Suitable prospect; Definition to preserve: The service, geography, and fit rules used at that time

• Baseline item: Appointment; Definition to preserve: Requested, confirmed, attended, or another specific status

• Baseline item: Sale; Definition to preserve: The source record and business event used to count it

• Baseline item: Marketing cost; Definition to preserve: Which media, provider, and outside costs are included

A form submission is not necessarily a new prospect. A meeting request is not necessarily an attended conversation. If the old records do not make those distinctions, preserve the limitation instead of estimating a precise historical figure that the evidence cannot support.

Keep the business conditions with the export

Record which services were available, any offer running, and whether staff capacity changed. Note known tracking problems and meaningful outages. A quiet period caused by a planned closure should not be presented as a normal month that the next provider is expected to beat.

Choose a reference period that represents the decision you are making. For seasonal businesses, one immediately preceding month may be a poor comparison. For a new service, there may be no historical equivalent. State the reason for the chosen period and avoid quietly replacing it when later results become inconvenient.

Build a small, reviewable evidence pack

Save a dated export from the relevant business systems, a snapshot of the active offer, the current service pages, and a short list of known weaknesses. Include the owner of each source. Do not collect unrelated customer details simply because an export allows it; the pack needs enough information to evaluate the marketing assignment.

Have the provider repeat back the definitions before changes begin. If it plans to count a different event, record the change separately. You can show the old and new measures side by side for a transition period rather than drawing a continuous line through numbers that mean different things.

Google Analytics distinguishes events, key events, and advertising conversions. Use the actual definitions in your setup. Analytics annotations can also mark a launch or another important event so later readers have context.

Agree on what the baseline cannot tell you

A before-and-after comparison can be useful without proving that the provider caused every change. Demand, pricing, service availability, and sales handling can also change. The baseline gives you a stable record to investigate those possibilities. It should not be used to manufacture a success story from a convenient pair of dates.

For reporting decisions after launch, read our marketing reporting guide. For comparing buying paths with the same outcome definitions, see measuring qualified inquiries from a landing page.

Escencion offers managed AI agents for website, search, ads, and inquiry qualification and booking. Build Your AI Agents with your starting evidence so the first assignment has a clear reference point.