Potbelly came into 2020 with a lunch strategy built on survey data, and then dining rooms closed. We rebuilt the message around what people at home were trying to avoid and ran it as one campaign across every way to order. Digital grew from 12.8% to 35.8% of sales while revenue grew 30% in 2021 and another 19% in 2022.


The result
12.8% to 35.8%
Digital share of sales, Q4 2019 to Q4 2020, from Potbelly's earnings release. Revenue $291.3M (2020) to $380.1M (2021) to $452.0M (2022).
“We are also enjoying early benefits from our strategic initiatives that are evident in our results, namely the benefits of our third-party delivery differential pricing and our targeted scaled media outreach.”
Potbelly had a strategy from an earlier engagement, built on survey data: people decide what to eat about 30 minutes before lunch, so the ads should run at lunchtime, to fans of other sandwich chains, under the line "You deserve a lunch break."
The brand's best customers came in about four times a year. Fans of the chains Potbelly competed with came in several times a month. Potbelly was a craving or a building people happened to pass, which meant it was rarely the first place they thought of.
Then the pandemic closed dining rooms. A plan built entirely on walk-in lunch traffic had no customer to serve, and digital orders, 12.8% of sales in Q4 2019, had to carry a public company with hundreds of locations.
Spotify, Pandora, programmatic, Facebook and Google each ran their own message, and the creative leaned so hard on shakes and cookies that some people read Potbelly as a bake shop. We paused the audio and programmatic spend and rebuilt from Facebook alone. When we brought channels back one at a time, Spotify and Pandora never earned their place: removing them made no measurable difference to sales or foot traffic.
The survey's timing assumption did not hold either. The lunch-break control cost $7 per walk-in. Easing the lunchtime restriction alone brought that under the client's $5 target, because food shown earlier in the day builds an appetite instead of catching a decision already made.
The biggest finding came after the dining rooms closed. We tested emotional reasons to order against the control, on the same video and the same images, changing only the words. Home cooking had become a source of pride online, and it was wearing thin. The friction at home sat in the cleanup, which caused arguments between partners and between parents and kids. "Tonight, let us do the dishes" returned 12 dollars for every dollar of ad spend, against 2 for the control.
The fix had to work while the business changed underneath it. It had to sell dinner as well as lunch and carry a national brand down to each neighborhood. It also had to count every order, whether it came through the app, the website, a delivery partner or a register.
And it had to answer to Potbelly's leadership in revenue. The CMO needed numbers the board would accept, which ruled out judging media on what each platform reported about itself.
We unified the data first. Online orders, mobile app orders, Facebook and Google data, and point-of-sale orders from every store went into one view, and the CMO presented media performance to the board from those dashboards. We also tied phone signals near each shop to in-store purchases. That method is directional, so we used it with holdout tests, never alone.
We localized a national brand. Every location got its own campaign, with drive-thru shops flagged, and the ads named landmarks instead of distances: the Potbelly next to a building people knew, in place of "2.3 miles away." People find places by landmarks, and the ads read like a neighborhood shop.
When the dining rooms closed, we rebuilt the campaign from the existing footage in two weeks, around online ordering and the new delivery partnership. Then we kept testing reasons to order and moved budget to whichever message beat the control, running it as one campaign wherever people chose to buy. Cost per walk-in fell to $2 while the average order grew from $11 or $12 to $24.
Potbelly kept a single campaign with one message sequence across every channel, built on a per-location targeting structure. The reporting tied media spend to orders in every channel, including the stores.
The program kept running after our engagement ended in mid-2021 and carried the business through 2022.
Non-catering digital sales grew from 12.8% of total sales in Q4 2019 to 35.8% in Q4 2020, according to Potbelly's earnings release.
Potbelly reported revenue of $291.3M in 2020 and $380.1M in 2021, up 30%. After our engagement ended in mid-2021, the program Potbelly kept running carried into 2022, when Potbelly reported $452.0M, up another 19%, and net income moved from a $23.8M loss to a $4.3M profit.
In the Q2 2021 earnings release, Potbelly's CEO credited "our targeted scaled media outreach" among the strategic initiatives showing early benefits.
Key takeaway
Surveys tell you what people say they do. Test the reason to buy in market, on the same creative, before you put the budget behind it.
Industries: Food, beverage and restaurants
Find problem areas today with the Market Health Analysis. It's $350, credited toward your Diagnostic.
Get your Market Health Analysis