A restaurant promotion usually goes out to the whole town, all week, for the place in general. Then the owner checks the covers and finds that the weekend was full anyway and the early week was not. Rent is lost in the quiet shifts. Campaign work for an independent dining room is not about reach. It is about moving demand into the shifts that still have open tables.
Why do my restaurant promotions fill Saturday and leave Tuesday empty?
Because the campaign was aimed at the whole town with no shift attached. Weekend demand mostly exists already, so paying to reach people who were coming anyway buys covers you would have sold for free. A promotion only earns its cost when it moves a booking into a shift with open tables.
Before you spend anything, work this out on paper.
- Write down the covers you can physically serve in each shift, given your tables and the staffing you schedule that night.
- Write down what you actually served in that same shift last month.
- The gap between the two is the only thing a campaign can sell. Where there is no gap, a campaign discounts guests who were going to pay full price.
- Pick one shift with a real gap and build everything around that shift alone.
Adam Erhart calls the general version of this the 80/20 rule and describes the goal as finding the "20% of things that are giving you 80% of your results". In a dining room that share is a shift on the calendar rather than a channel in a media plan. Your own cover counts by shift, not a benchmark from anywhere else, tell you which shift it is.
Should I advertise my whole menu or just one dish?
Advertise one thing and keep cooking everything. A promotion that lists your entire menu tells nobody what to order tonight. Pick one dish, one deal or one time slot per campaign, with a single call to action. The kitchen keeps its full range. The ad does not.
Choose the dish on three filters you already hold the data for.
- Margin. Check your own plate costing. A dish with a thin margin turns extra volume into a loss.
- Prep reality. Use something already in your mise en place. An offer built on an ingredient bought in specially leaves you holding spoilage if the campaign underperforms.
- Line speed. If the dish slows the pass on a busy service, it is the wrong one to promote.
Erhart opens his video with the same instruction for business owners in general: "Stop trying to sell everything to everyone and start selling just one really good thing to the right people." His reasoning is that "People tend to remember specialists, not generalists." and that "one really good offer beats 10 mediocre ones".
Where this parts company with the video. Erhart tells owners to take everything outside the core offer and "bundle it into different packages or cut it completely". A kitchen cannot cut its menu down to one plate. The one offer rule belongs to the campaign, not to the menu.
How do I get repeat guests without paying delivery app commissions?
By collecting contact details from guests who already ate here, so you can invite them back directly. A delivery app keeps the guest data and takes a commission you can read on your own statements. A list you own lets you fill a slow Tuesday at full menu price, on the day you choose.
Capture that does not depend on staff memory, because front of house turns over.
- A QR code printed on the receipt and on the table tent
- A wifi login that asks for an email
- The booking and waitlist forms you already run
- A birthday club, which gives you a reason to make contact on a date you pick
Erhart's argument for an owned list is that everything else is rented. "Social media reach can disappear overnight and it often does", and paid ads "stop the moment you stop spending."
He also cites a return of "38 to 44 bucks for every dollar spent" on email, which he attributes to studies he does not name, measured across businesses in general and not restaurants. Do not carry that number into your own forecast. Use it as a reason to run one send to your list and count the covers that come back.
One adjustment for this niche. His cadence advice is "I recommend emailing one to three times per week at least." A list built from diners is not a list of subscribers, so test that frequency against your own unsubscribe rate before adopting it.
Where should a small restaurant show up online?
Show up where people who can physically reach your dining room already are, then stay there. For one location that usually means your Google Business Profile plus a single social account you can actually keep current. Running five accounts badly costs the one thing service will not give back, which is your attention.
Erhart's rule for choosing is blunt: "Pick one platform, go all in, then get really good at it." His pointer for local businesses is direct: "If you serve an audience over 40 years old, or you serve local markets, Facebook still wins."
On search he is emphatic: "If you serve a specific city or region or small geographic area, then local SEO is one of the highest ROI channels out there." He says that setting up a Google business profile, buying citations and publishing local content lets you "dominate the Google map pack and get calls and visits and leads" without spending on ads.
For a dining room that means the unglamorous work first. Correct hours with holiday hours included, a current menu, photographs of the actual room and the actual plates, and a habit of replying to reviews. You can measure your own catchment yourself by checking where your last hundred bookings or orders came from, then keeping the radius inside it.
What does a simple marketing funnel look like for a restaurant?
Four steps you build once and leave running. An ad or post carrying one offer, a page that takes a booking or a contact detail, a short email or text sequence, and a follow up that asks for a review. No step should need you at a laptop during service.
Erhart's most useful correction here is one line: "Your website's homepage is not a funnel." Traffic from a Tuesday offer that lands on a homepage full of links, opening hours and a photo gallery has nowhere obvious to book.
He also warns against waiting for the perfect build: "A simple funnel that's live is always going to beat the perfect funnel that you never launch." On judging results, he says to give a funnel somewhere around a thousand clicks or views before deciding whether it works. A single dining room may take a while to reach that, so plan to leave it running rather than switching it off after one slow week.
What tools do small restaurants use to run their own ads?
There is no single answer, because each tool covers a different part of the job. The useful question is which part you are missing right now. The table below lists what each one does, what it changes inside a restaurant, where it stops, and what it asks you to know before it works.
| Tool | What it does | What it changes inside a restaurant | Where it stops | What it asks you to know |
|---|---|---|---|---|
| Meta Ads Manager | Builds and runs ads across Facebook and Instagram | Lets you point a Tuesday offer at a location radius and at chosen times instead of at everyone, all week | It will not tell you which shift needs filling, and the result depends on the creative you supply | You set the audience, the objective, the budget and the creative yourself |
| Google Ads | Runs ads on search and on maps | Puts you in front of someone typing a food search near you while they are still choosing | You pay for clicks, not for bookings, and it does not fix an incomplete Google Business Profile | You choose keywords, match types and bids yourself |
| Canva | Template based design tool | Menu boards, offer graphics and story posts without hiring a designer | Produces the artwork only. It does not target, publish to ad platforms or measure anything | Design basics. There is no advertising side to learn |
| Mailchimp | Email marketing and automation | Sends the offer to the guest list you own, on the day you picked to fill | You still have to collect the list and write the emails, and the list is only as good as your capture habit | Enough to build a list, write a send and read an open and click report |
| ManyChat | Automated messaging on Instagram, Messenger and WhatsApp | Answers repeat questions about hours, bookings and parking without pulling someone off the floor | Every reply has to be written and mapped in advance, and it stays subject to each platform's messaging rules | You write the replies and map the conversation flow yourself |
| SaleADS.ai | AI software that creates and launches advertising campaigns on Meta, Google and TikTok for business owners, with no design or advertising expertise required | The campaign build happens inside the software instead of inside an ads account | It has no view of your covers, your plate margins or which shift is short, and those are the decisions this article puts first. This site has not tested the claim made in its description | Its own description states that no design or advertising expertise is required |
SaleADS.ai is the product of the company that publishes this site. It appears in this table as one option among several and is not a recommendation. Which of these is worth your money depends on which part of the job you are missing, and only your own numbers answer that.
The descriptions above come from each vendor's own product information. Erhart does not evaluate any of these tools in the video. He ends on the businesses that are winning right now: "They're simply executing the fundamentals better than everyone else."
Where does this information come from?
Every quoted sentence on this page comes from one video, 5 BEST Marketing Strategies For Small Business 2026, published by Adam Erhart and running 18 minutes. Each quote links to the block of the transcript where it is spoken. The sections below separate what the video says from what this article adds for restaurants.
Source: "5 BEST Marketing Strategies For Small Business 2026 (PROVEN & PROFITABLE)", Adam Erhart, 18 minutes.
What comes from the video: the one core offer principle, the 80/20 framing, the one platform rule, his guidance that local markets still do well on Facebook, the local SEO argument, the case for an email list you own, the four step funnel structure, the one to three emails per week cadence, and the rough threshold of a thousand clicks before judging a funnel.
Figures that are his and are not restaurant data: the 38 to 44 dollars returned per dollar spent on email, which he attributes to studies he never names, and the thousand click threshold. Both come from his general small business work. Neither is a restaurant benchmark. This article quotes no restaurant sector statistics, because it has none it can source.
What this article adds, with no external verification: the capacity gap exercise, the margin, prep and line speed filters for choosing the dish to promote, the staff turnover argument for automated capture, and the delivery app framing. That is reasoning applied to a dining room, not a finding.
Where this article disagrees with the video: Erhart tells owners to bundle or cut everything outside the core offer, which a kitchen cannot do to its menu. He recommends one to three emails per week, which this article treats as a hypothesis to test against your own unsubscribe rate, because a list of past diners is not a list of subscribers.
Not used from the video: the segment in which Erhart recommends a paid SEO vendor, along with the vendor statistics he reads out in it.