DelSur Empanadas, published reels

A sample of the twenty-seven reels published for DelSur between March and August of 2026. Every one produced from concept through filming direction, editing, captions, and scheduling.

The Challenge

DelSur Empanadas has been the Twin Cities’ first Argentine empanada restaurant since 2018, founded by Diego Montero and Nico Nikolov, who have been sharing Argentine food and culture in Minnesota since they launched their food truck in 2014. When I came on in February 2026, the restaurant already had everything a marketer hopes to be handed: tango nights, a food truck, a second location at Malcolm Yards, and a dining room that fills every time Argentina plays. What it did not have was a plan for any of it.

The accounts were running on stories that reached between 115 and 288 people and then disappeared within a day. Instagram reached 685 people that month and produced 78 interactions. The Facebook page, which had not been looked at in some time, drew 424 viewers, 21 interactions, and a single link click across the entire month.

I also began without the basics. There was no back end access, no communications manual, no hex colors, no logo files, and no fonts. I asked for all of it in our first meeting, and I received Meta Business Suite access on February 24.

The real problem was never that DelSur needed to post more often. A restaurant with genuine cultural standing in this city was reaching roughly the same number of people as a personal account, and the one summer when the whole metro would care about Argentine football was three months away.

The Approach

I built the strategy around four decisions. I want to explain the reasoning behind each one, because the reasoning is the part that transfers to any other client.

Change the format before increasing the volume. Stories were quick to produce and gone within a day. Reels crossposted to Instagram and Facebook returned far more for the same hour of production, and Facebook was sitting entirely unused. The same piece of content on a second platform costs nothing extra to make, which made it the highest return decision available to me.

Put people on screen, not only food. Looking back through what had already performed, the posts with staff and customers in them consistently outreached the ones without. Food photography is interchangeable from one restaurant to the next. Diego at the grill is not. So I built recurring series around the team, including rapid fire questions with the kitchen while they worked and short profiles of the front of house staff.

Treat neighboring businesses as distribution. The single best performing post in February was a collaboration with the brewery next door. Partner accounts were free reach that nobody was asking for, so collaborations with Unmapped Brewing and Malcolm Yards became a deliberate part of the plan rather than a happy accident.

Let events carry the calendar. A restaurant with tango nights, a food truck, live music, and a World Cup approaching has no shortage of reasons to be talked about. I built the schedule around those moments instead of around another photograph of an empanada.

Events became the spine of the calendar. Tango nights, live music, food truck season, and in-house table promotion.

The Process

February and March, building the system. I created an editorial calendar with post identifiers, publication dates, on screen hooks, captions, and line by line subtitle scripts. Each reel received a shot recipe, meaning a numbered list of up to eight shots with audio notes, so that filming could happen on a day when I was not on site. A story bank and a production tracker went alongside it.

March and April, refining. The logo moved into the first three seconds of every reel. The personality series launched. When the owners commissioned outside brand research, I held off on further format changes until it arrived, because guessing ahead of information you have already paid for is wasteful.

Then the volume. Between March and August I produced and published twenty-seven reels for DelSur, alongside forty-two for Tonka Pizza in the same period. Every reel and every image in that run is mine, from concept through filming direction, editing, captions, and scheduling.

April through July, events. Food truck season, Sunday brunch, take and bake frozen empanadas, Tango Night on Eventbrite, and the World Cup.

The Flagship Campaign, The World Cup

Argentina’s run was the moment DelSur’s position was worth the most, and it was the only thing on the calendar that could not be rescheduled.

The campaign opened on June 4 with a Spanish language poster carrying all three group stage match dates, converted to Minnesota time. Reels ran through the tournament. The events went onto Eventbrite and Facebook. The owners arranged an overflow venue with the brewery down the block.

Fox 9 responded to the release and interviewed a co-owner on camera at the party. El Minnesota de Hoy, a Spanish language outlet from the Latino list, covered the event.

The Outcome

Instagram reach grew about 6x. Facebook, which had been dormant, grew roughly 19x.

Instagram reach grew about six times over the period. Facebook, which had been dormant, grew roughly nineteen.

Three findings matter more than the totals.

The steepest jump came in March, not in July. Instagram reach went from 685 to 4,000 and Facebook viewers from 424 to 5,000 in the first full month of the new format. That happened before the World Cup existed as a story, which tells me the change came from the system rather than from the season.

Facebook turned out to be the larger opportunity. It had been left dormant, and over six months its viewers grew roughly nineteen times while link clicks went from one to 197 and visits from 550 to 1,700. Crossposting cost me nothing in production time and returned more than anything else I did.

The press campaign produced two placements from a restaurant that had never pitched, along with a full house and an overflow venue that was needed and used.

Facebook Engagement

From one click in February to 197 in July, on a page nobody was using.

The clearest single measure of the crossposting decision: one link click in February, 197 in July.

A packed dining room at DelSur watching the World Cup final on July 19th. The restaurant filled and kept filling, and the crowd carried over in to the overflow venue next door.

What I Would Do Differently

May was the weakest month on both platforms, and I do not have a flattering explanation for it. June and July were carried in part by the World Cup, and while real work went into being ready to catch that attention, the tournament itself was not my doing. The account also started from a low enough base that the early multiples look more dramatic than the underlying change. I would rather say all of that plainly than let a reader discover it on their own.