The short answer
MMM budget forecasts should use the store network expected during the planning period, including opening dates and assumptions about new stores' sales. Separate the effect of that changing network from the effect of a media budget change. Compare media plans under the same store assumptions, and make any baseline adjustment explicit.
This article is part of Asked by Marketers, a series answering real questions from marketing leaders.
Sellforte's team holds more than 1,450 meetings each year with marketing leaders in Ecommerce and Retail about Marketing Mix Modeling and incrementality testing. Each week, we anonymize at least one question from those conversations and answer it in depth, based on what marketers are actually struggling with, not what keyword tools suggest. About the series →
Why marketers ask this
A retailer planning next quarter's advertising may have more stores than it had in the historical comparison period. The media team sees a higher sales target and needs to understand how much advertising will contribute. Finance also needs a forecast that reflects the expansion already planned.
When last year's sales and advertising provide the starting point, check which store network the forecast describes. A higher total sales forecast can reflect more places to buy, a stronger media plan, or both. The budget discussion needs to separate those effects.
Does the historical MMM already account for store openings?
A model can account for historical store openings without automatically knowing the future opening plan. Check how changes in the store network enter both the historical model and the forecast.
In Marketing Mix Modeling, store availability is one of the business conditions that can help explain sales alongside advertising. Depending on the setup, a growing network may appear in a baseline trend, in an explicit store-related component, or through a model of average sales per store that is converted back into total sales.
Those approaches have different implications for planning. A historical baseline may reflect the stores that were open then, while a future scenario requires an assumption about the stores that will be trading during its dates. Ask whether the forecast has already incorporated that change before adding an adjustment. Otherwise, the same expansion could be counted twice.
Check the units as well. Total network sales and average sales per store answer different questions. If the model works with a per-store outcome, confirm how store counts and opening dates are used to produce the total forecast. A rise in total sales with a larger network does not, by itself, mean that existing stores or advertising have become more productive.
What store assumptions should the forecast use?
Start with the planned store calendar and a defensible estimate of its sales effect. A count at the end of the quarter is insufficient when stores open at different points during it.
Record where each opening falls in the model's regions and when it begins trading. Include closures that affect the comparison. Then agree with the team responsible for the commercial forecast how new locations are expected to perform during the planning period. Treat their ramp-up as an assumption to examine; do not assume every new store immediately produces the average sales of an established store.
The forecast should also state whether expected new-store sales include any transfer of sales from existing locations. If that effect is unknown, show a range of plausible network assumptions instead of treating the new location's entire sales forecast as additional business.
In some setups, these assumptions can enter through store-specific inputs. In others, the team will need an explicit adjustment to expected baseline sales. Confirm the supported method for your model. A generic percentage increase should be documented as a planning assumption, rather than presented as a store-opening effect the MMM has independently estimated.
How should we compare media plans during expansion?
Hold the future store network constant while comparing alternative media plans. That gives the budget comparison a common set of business conditions.
First, establish the expected network and other planning assumptions, including prices and promotions. Then compare the current media plan with the proposed allocation for the same period. Keep a separate comparison that shows how the changing network affects the forecast. Choosing last year's matching weeks can help with seasonal timing, but it does not make last year's store footprint suitable for this year.
Opening advertising also needs its own treatment. The opportunity to buy from a new store and the advertising that promotes it are separate drivers, although their effects can be difficult to distinguish when they begin together. Record opening campaigns in the media plan and check that the commercial baseline assumption does not already include the same advertising uplift.
More stores do not establish that every channel deserves the same percentage budget increase. The model's media estimates, the regions affected, and the budgets the team can execute still need to support the proposed plan. Where new locations differ substantially from the historical network, report that limitation in the forecast.
How this looks in practice
Consider a hypothetical retailer with 90 existing stores and 10 openings planned during the next quarter. The following figures are illustrative scenario assumptions, not observed results. For simplicity, all non-media sales contributions are combined in one column. Prices, promotions, and the forecast period stay the same across the scenarios.
| Scenario | Non-media sales | Media effect | Total sales |
|---|---|---|---|
| Existing network, current media plan | 8.0 | 2.0 | 10.0 |
| Planned network, current media plan | 8.6 | 2.0 | 10.6 |
| Planned network, revised media plan | 8.6 | 2.3 | 10.9 |
The €8.6 million assumption comes from the retailer's hypothetical network forecast. It is not calculated by multiplying last year's sales by the percentage change in store count. It assumes the opening calendar and expected net sales from new stores have already been considered.
This example also holds media contribution at €2.0 million when only the network changes, to make the comparison easy to follow. A real forecast may estimate a different media contribution for the larger network; that relationship needs to be checked rather than assumed.
Under the stated assumptions, expansion adds €0.6 million to forecast sales before the media plan changes. The revised media plan adds another €0.3 million. Crediting advertising with the full €0.9 million difference would mix the two effects. Repeat the media comparison under a lower network-sales assumption if opening dates or early trading expectations remain uncertain.
How Sellforte helps
Sellforte's MMM and budget planning tools help teams compare media scenarios alongside baseline and promotion assumptions. For a changing store network, the modeling team can establish how openings enter the historical model and which future assumptions belong in the planning setup. Book a demo.
Authors
Lauri Potka is the Chief Operating Officer at Sellforte and has over 15 years of experience in Marketing Mix Modeling, marketing measurement, and media spend optimization. Before joining Sellforte, he worked as a management consultant at the Boston Consulting Group, advising some of the world's largest advertisers on data-driven marketing optimization. Follow Lauri on LinkedIn, where he is one of the leading voices in MMM and marketing measurement.
How should MMM account for returns before the return window closes?
Can MMM help us set a budget for a channel we've never used?
