Blizsocials

INSIGHTS

How to brief a social media agency so the work comes back right

What to bring to the first call, what to insist on in the contract, and the reporting you should expect every month.

Four members of the Blizsocials team outdoors in branded shirts against a clear blue sky

Most work that comes back wrong was briefed wrong, and usually not because the brief was too short. It was because it described the output instead of the outcome, twelve reels instead of the thing the twelve reels were supposed to achieve, which means the agency can deliver everything they promised and you can still be unhappy, and neither of you can say exactly why.

The short version

• Brief the outcome you want, not the list of things you want made.

• Bring what you sell, what you want changed, what you have tried, and who signs off.

• Own your files, your logins and your ad account, in writing.

• Judge the first month on the questions they ask, not the follower count.

What to bring to the first call

Five things. Written down, even roughly. Bring them and the first call stops being an interview and starts being work.

What you sell and who buys it. Not the mission statement, the actual thing money changes hands for and the actual person who hands it over. What you want to change: more enquiries, better enquiries, a different calibre of customer, a reputation problem you are tired of. Pick the one that matters most, you can have the others later.

What you have already tried, including what failed. This is the single most useful thing a client can give an agency and the thing most clients hold back, usually out of embarrassment. Nobody is judging the ₦200,000 you spent on the influencer who posted once. We just need to know so we do not spend your money learning it twice.

What you cannot say: regulatory limits, claims you are not allowed to make, subjects the brand stays out of. And who signs off, and how quickly. An account with a four day approval loop cannot post daily. That is fine, but it needs to be known on day one, not discovered in week three.

Describe the outcome, not the output

"We need twelve reels a month" is a shopping list. It tells us what to make and nothing about what it is for, which is how work gets delivered in full and lands wrong.

"We want more of the customers who book the full package rather than the trial" is a brief. It can be argued with, planned against and measured. It might produce twelve reels. It might produce four proper films and a rebuilt profile, and you would probably prefer the second one.

Nobody can hit a target you have not described.

What belongs in the contract instead

Four things, and a decent agency offers all four before you have to ask. Named deliverables in numbers, with dates. Ownership of files, footage, logins and the ad account, all of it yours, in writing, and you should be able to walk away with the lot. What happens to your ad budget: held separately, reported separately, spent in your name. And a notice period you can actually use, long enough that the work gets a fair run and short enough that you are not trapped with people you have stopped trusting.

What to expect every month

A report you can read in five minutes that answers three questions: what moved, what did not, and what changes next month because of it. A screenshot of a dashboard is not a report. A dashboard is data, a report is a decision.

And if the month went badly, that belongs in the report too. An agency that only sends you the good months is managing you rather than the account, and by the time you notice you will have paid for two quarters of it.

Your half of the bargain

The accounts that work best are the ones where the client answers questions within a day, sends us the raw material only they have (the new stock, the finished project, the customer who said something worth repeating) and resists the urge to rewrite every caption into the voice of a press release.

The biggest single cause of average work is a slow approval loop. Content is perishable. A post approved eleven days late is not a late post, it is a different post, and usually a worse one.

How to tell in month one

Not by the follower count, which will do whatever it wants for the first six weeks. Look at this instead. Are they asking you questions you had not thought about? Does the work look recognisably like you rather than recognisably like them? Have they told you something you did not want to hear? And did the first report actually change the plan, or just describe it?

Four yeses and you are in good hands. Four noes and month two will look exactly like month one, which is the real thing to be afraid of.