Or give it to the spreadsheet guy?
Nobody in a restaurant group is choosing between me and another AI trainer. The real decision is whether to spend money at all, or hand it to the manager who is already good with spreadsheets.
Quite often that person is the right answer. They are cheaper, they know your business, and they are still there next year.
The case for bringing somebody in is narrower than the people selling it admit: you are buying compressed time and a shorter list of wrong turns, not access to something your own team could not eventually work out.
When your own person is the better call
The problem is contained. One report, one store, one process. That is a good first project for somebody internal and a bad reason to bring in an outsider.
They have actual protected time. Not enthusiasm, not evenings. Hours in daylight that somebody senior has agreed are for this and will defend when service gets busy.
You want it to stick. Anything owned by a person in the building outlasts anything owned by a visitor. That is true of every consultant including me, and any consultant who tells you otherwise is selling a retainer.
You are above about twenty locations. At that size the work is closer to a real job. Hire for it, and use somebody outside to brief them rather than to do it.
The spreadsheet guy is usually the best thing in the group. Nobody has ever asked him to show anyone.
When bringing somebody in earns its money
You need several locations doing the same thing. An internal person can build something excellent for their own store and has no authority to make store nine adopt it. Consistency across a group is a different job from building a tool.
The learning curve is the cost. Six weeks of a manager’s Fridays is real money and real Fridays. Compressing that into a day is most of what you are actually buying.
Nobody internal has the time and everyone says they do. This is the most common one. The project starts, service gets busy, and eight weeks later it is at forty percent and nobody wants to say so.
You want the wrong turns skipped. Not because they are hard, but because finding out which vendor tool already does the job takes an afternoon if you have done it before and three weeks if you have not.
The version most groups should buy
A day to set direction and compress the learning, then your own person owning it from there with somebody to call when they get stuck. It costs less than a full engagement and survives better than a purely internal project, because the thing that kills internal projects is the blank page rather than the ability.
If you want to test that cheaply, an hour on the phone is $500 and will usually tell you which of these three you are. I would rather send you away with a brief for your own person than sell you a day you did not need.
More on working with a group
- Private AI training for restaurant groups
A day with your own team, on your own numbers.
- What a private engagement costs
From $5,000, and exactly what moves the number.
- Rolling out across multiple locations
When every GM has invented their own version.
- Implementation and building
When the answer is that somebody has to build it.
- Public workshops
The cheaper way in, if you have one or two to train.
Questions people actually ask
- We have someone good with spreadsheets. Should we just use them?
- Often yes. If they have genuine time - not 'time' that means evenings - and the problem is contained, they are cheaper, they know your business, and they are still there in a year. The failure mode is not capability, it is that nobody protects the hours and it becomes a side project that never finishes.
- What is the real cost of doing it internally?
- The salaried hours, which are not free even though they are already being paid. A manager at a loaded $30 an hour spending six weeks of Fridays on this is roughly $1,440 of time, plus whatever those Fridays were for.
- Should we hire a dedicated person?
- Below about twenty locations, almost never. The job is not full time and you will end up with someone underused and expensive. Above that it starts to make sense, and at that point the useful thing I can do is help you brief them rather than take the work.
- What about our POS vendor's AI features?
- Use them. Seriously - you already pay for them and they are integrated with your data, which is more than an outside tool can say. Finding out what your vendors already give you is usually the first hour of any session I run.
- Can we do both?
- That is the common answer. Bring someone in to compress the learning curve and set the direction, then hand the ongoing work to your own person who is now not starting from a blank page. It costs less than either extreme.
- How do we brief an internal person properly?
- Give them one process, protected hours in daylight, a named stakeholder who wants the outcome, and a number they are trying to move. Missing any one of those four is why most internal projects stall, and none of them cost money.
Still deciding?
Take the free check first. Eight questions, no email, and you get the two things I’d fix in your building before I ask you for anything.