The Portfolio Averaging Problem
When a multi-location brand runs a marketing campaign, the results are almost always reported as a portfolio average. Total impressions. Total clicks. Total visits. Average CAC. Average ROAS. Those aggregate numbers then drive decisions about future spending.
The problem with averages is that they obscure the actual distribution of performance. A campaign that produces strong results at four locations and weak results at six others will show an "average" result that accurately describes none of the ten locations. Decisions made from averages are decisions made from fiction.
The truth is that every location in a multi-unit brand is a different competitive situation. The customers are different. The competition is different. The traffic patterns are different. And the performance of any given marketing campaign will reflect those differences — whether or not the reporting does.
Why Location Performance Varies
Ask any multi-unit operator and they will tell you: the stores that struggle do not struggle randomly. There are patterns. The underperformers are usually dealing with one or more of the following:
- Increased local competition. A new competitor opened nearby and drew away customers who used to be regulars.
- Demographic shift. The surrounding population changed — moved away, aged, or changed spending patterns — and the location's customer base did not replace itself.
- Execution issues. Service quality, staffing, or operational consistency is driving churn that marketing cannot fix.
- Location disadvantage. Traffic patterns changed — a road closure, a new development, a parking situation — and visibility or accessibility declined.
- Underdeveloped market. The location is in an area where brand awareness is simply lower than in more established markets.
Each of these situations requires a different response. Running the same campaign that works for the brand's strongest locations will not address any of them.
The Case for Location-Level Marketing Intelligence
The most actionable thing a multi-location brand can do is understand which locations are underperforming and why — at the marketing level. That means moving from portfolio-level reporting to location-level reporting.
| Portfolio Reporting | Location-Level Reporting |
|---|---|
| Average CAC across all stores | CAC by individual location |
| Total campaign visits | New customer visits per store |
| Blended ROAS | ROAS by location and market |
| One creative/message for all | Tailored messaging by competitive situation |
| Same budget allocation everywhere | Budget weighted to opportunity |
NXTeck is designed to operate at the location level by default. Every campaign is measured by individual store — not by portfolio average. That makes it possible to see not just that a campaign worked, but where it worked, and what the performance differential looks like across locations.
How to Think About Location-Level Budget Allocation
Once you have location-level performance data, budget allocation becomes a more deliberate decision. A few frameworks that work in practice:
Invest where traffic has softened, not just where it is strongest. The most counterintuitive principle in location marketing is that your marketing budget often produces the highest incremental return at underperforming locations — not the flagship stores. If a location is already at capacity, a campaign produces little incremental value. If a location has significant untapped potential, the same campaign spend can have a much larger impact.
Set location-level targets, not portfolio targets. A target of "increase traffic by 10%" means very different things at a store doing 800 visits per week versus one doing 200. Express targets in absolute new customer terms — number of new customers acquired — by location.
Separate marketing problems from operations problems. Some underperforming locations have a marketing problem: not enough potential customers know about them or are being reached. Others have an operations problem: customers visit once and do not return. Marketing investment in an operations problem is wasted. Location-level attribution data helps identify which is which.
The NXTeck Pilot Framework Applied to Location Selection
NXTeck pilots are almost always structured around location selection first. The standard recommendation is to start with locations where traffic has slowed or where competitive pressure has increased — for two reasons.
First, these are the locations where incremental marketing impact is most measurable. If a store's traffic was declining and it reverses during a NXTeck campaign, the attribution is cleaner than at a store where traffic was already strong.
Second, if NXTeck can produce results at the hardest locations — the ones under competitive pressure, with softening traffic — the case for broader expansion is obvious. The pilot proves the platform where the stakes are highest.
We test where results actually matter. If the campaign works at your hardest locations, you already know what to do next.
Practical Steps Toward Location-Level Marketing
- Rank your locations by trailing 90-day traffic trend. Identify the bottom quartile by traffic growth — these are your primary candidates for acquisition-focused campaigns.
- For each underperforming location, diagnose whether the problem is marketing (low awareness, low reach) or operations (low repeat rate, poor conversion). Attribution data helps here.
- Design campaigns that address the specific location's competitive situation — not a generic brand campaign.
- Report results by location, not by portfolio average. Require your media partners to provide location-level attribution.
- Adjust budget allocation quarterly based on location-level performance data.
The Bottom Line
One campaign rarely fits all locations. The brands that understand this build location-level marketing intelligence into their operating rhythm — knowing which stores are underperforming, why, and what specific actions are most likely to move the needle at each one.
The ones that run portfolio campaigns and measure portfolio averages are making decisions from numbers that describe their business inaccurately — and spending money on strategies that may be right for some locations and wrong for others.
