Select Page

April 4, 2017

The website MathIsFun defines “permutations” as all possible arrangements of a collection of things, where the order is important.

Let’s say you have a chain of 500 hardware stores, some company owned, some franchised, some of those franchisees part of ad-buying groups, a corporate Facebook presence. You want to increase store visits and the resulting purchases. You are competing with the big boxes of Home Depot and Lowe’s and with the omnipresent Amazon. But, your customers can still be highly motivated by convenience and urgency. When a homeowner needs a toilet float, chances are he or she needs it right now and isn’t in the mood for a time-consuming shopping excursion. You have a customer base and are important to your local community.

You then make the wise decision to allocate part of your marketing budget to Facebook advertising. You do that not because it’s trendy but because it is very efficient, precisely targeted, and delivers better results than old-style direct mail or newspaper inserts.

Now is where, without the proper assistance, permutations throw into your plans a proverbial wrench, as opposed to a wrench off one of your store shelves.

Why? Consider the life cycle of running a flight of Facebook ads for a collection of 500 stores, where for each one the messaging may need to be a bit different and where all will be set to their individual geographic targets.

Full Lifecycle

 

The life cycle steps are these:

Planning – What type of promotion do you want to run? How will it vary by city and region as seasonal changes occur according to different rhythms? Snow shovels might still be of interest in the northernmost part of your chain while pool chemicals are in demand in the southern areas.

Buying – You must select from a long list of criteria the types of customers you want to reach and during what time span you want to be viewed by them. You can literally drop a pin on the neighborhoods most likely to patronize your stores – the homeowners as opposed to renters, the better incomes, and the areas where neighborhoods are well kept. You know what you are spending, but you won’t know in advance what will be your CPM.

Proofing – Every ad for every location must be proofed. There will be errors on the first pass. Links can be off by a character and be useless. Typos will sneak in. You need to double check that the correct creative is going to each targeted audience.

Moderation – Comments on a Facebook post can build credibility and magnify your impact, but you’ll need to excise about 10% of them. Not everyone is respectful or accurate in making comments. You’ll get your share of undesirables.

Reporting – You’ll want some immediate reporting from your stores as to ad results. What were the actual CPM costs? Was there a lift in sales as a result of your ads? Where and with what creative elements did you get the best returns?

Optimizing – With the data you have collected, you will want to optimize your next flight of ads to overweight what you know works best. You can A/B test all aspects – from demographics to creative to specific offers to timing. Let the data be your guide; it’s more powerful than your intuition.

Billing and Reconciliation – If you are sharing costs with franchisees or their group buying arrangements, you’ve got to split the tab at the end of the campaign. That’s no small challenge to handle fairly and expeditiously.

Summary – Once the campaign has run its course, what is your conclusion? Once again, you can let the data tell you. There’s no guessing, just evaluation of the facts.

Consider all these eight steps. Multiply them across 500 stores times perhaps dozens of creative concepts times 100’s of ad selection criteria times multiple iterations for optimizing and you’ve easily got several thousand individual decisions and calculations to make. That’s well beyond practical handling by a small team with spreadsheets. It requires a data science driven platform that allows one or two ad operations experts to be sure you are making your Facebook spend wisely.