What to approve, and what to send back

Introduction

Nothing your operators make gets published until you say yes. That is the point of the whole arrangement, and it is also the part that makes people nervous. Once you are the one saying yes, a mistake feels like your mistake, and the natural response is to read every sentence twice, then leave it until tomorrow, then find eleven items waiting for you on Friday.

There are two ways to get this wrong, and they pull in opposite directions. You can wave everything through without looking, in which case your website ends up saying things about your business that are not true. Or you can hold everything until you feel certain, in which case the work stops, because none of it moves without you. The Content Writer can write, the Social Media Manager can prepare a post for your accounts, the E-commerce Manager can get a storefront change ready, and all of it sits still.

The way out is not to read harder. It is to read for the few things that actually matter, decide quickly on everything else, and be specific when you do send something back. This is a narrow job, and once you know how narrow it is, it takes minutes rather than evenings.

Three questions, in this order

You are not the editor. The Content Writer already handled the grammar, and the Web Designer already made the page look like a page. Your job is smaller than theirs, and it is the one job nobody else can do, because you are the only person who knows what is actually true about your business.

Read anything that reaches you with three questions, in this order:

1. Is it true? Every checkable fact: prices, hours, materials, ingredients, sizes, delivery windows, where you are, how someone reaches you, how long you have been doing this. If you did not give the team a number, treat that number as suspect until you confirm it.
2. Can you actually do it? Copy makes promises. If the page says orders ship in two days, then you ship in two days, every week, including the week you have the flu. Read every promise as something you are signing.
3. Does it sound like you? Read it out loud. If you would never say that sentence to a customer standing in front of you, it is not yours yet.

The order matters, because the first two are facts and the third is judgment. A false sentence has to be fixed before it goes anywhere. A true sentence that feels too formal for you is a preference, and preferences are worth working on over the coming weeks, not worth stopping the week for.

If something passes all three and still nags at you, name the nag before you send it back. "I don't like it" gives the team nothing to work with. "This sounds like a chain store, and I am one person in a kitchen" gives them something they can change.

Not everything deserves the same read

Your operators produce very different kinds of work, and reading a caption the way you read a product page is how a review habit collapses. Sort what arrives by one question: will a person act on this?

Read closely, every word. Anything a customer makes a decision on. Prices and shipping rules from the E-commerce Manager. Refunds, deposits, and anything about how money moves, which is where the Chief Financial Officer's work shows up in front of customers. Replies the Customer Support Agent has drafted to a real person who asked a real question. Product descriptions where the details are the product: allergens, dimensions, materials, care instructions, what is in the box.

Read once, at normal speed. Blog posts, the story on your about page, social posts from the Social Media Manager, the copy on a new page. Check the facts, listen for the voice, then move on. These are the pieces where a slightly awkward sentence is not worth a second round.

Skim, and trust the specialist. Page titles and meta descriptions from the SEO Specialist, alt text, layout changes from the Web Designer, the plumbing work from the Web Developer and the Webmaster, and the numbers the Data Analyst brings back. Glance to check that your business name and what you sell are described correctly, then let it go. You do not need an opinion about a heading tag.

One exception cuts across all three piles. Anything that names you personally, your home address, your phone number, or your family gets the close read no matter which operator wrote it.

Approve things you would have written differently

The most useful habit you can build is approving work that is correct but not what you would have written. You will do it regularly. It feels like lowering your standards. It is not. It is separating "wrong" from "different", and that is the skill that keeps the week from waiting on you.

How careful to be honestly depends on the piece. Here is what it depends on:

• Will someone act on it? A customer reads a delivery promise and plans around it. Nobody plans around an adjective.
• How hard is it to take back? A page can be rewritten. A post goes out on an account you connected yourself, so you can delete it. A reply that has landed in someone's inbox cannot be recalled.
• Is it a description or a commitment? "Made by hand in my kitchen" describes. "Ships within 24 hours" commits.
• Does it touch money? Prices, discounts, refund terms, and anything a payment attaches to get the slow read even when the wording is dull.

Push the reversible, low-stakes work through fast. If a caption is true, keeps a promise you can keep, and does not make you wince, approve it. You can tell the team you want a plainer tone next time without holding this one.

Save your attention for the sticky things: the homepage, the product pages, the shipping and returns policy, and your prices. Those get read by everyone and they set expectations you will be held to for a long time.

What to send back, every time

Some things are not judgment calls. Send these back without agonizing, even when everything around them is good.

• A number you never gave them. Any price, delivery time, quantity, year founded, team size, or measurement you did not supply. Correct it, or ask for the sentence without a number in it.
• A promise you have not agreed to. Same-day anything, free returns, replies within an hour, always in stock, satisfaction guaranteed. If you would not do it on your worst week, it does not get published.
• Credentials. Certified, licensed, insured, trained, award winning, member of. These words carry legal weight. If it is not on a document you hold, it does not go on your site.
• Claims about what the customer will get out of it. Anything promising a result, especially about health, safety, money, or legal matters. This is the category that causes real trouble.
• Other people's names and property. A competitor named in a comparison, a supplier named without permission, a photo or logo you do not own, a quote attributed to a customer who never said it.
• Money mechanics you have not set up. A discount code that does not exist, a subscription you do not offer, a payment method you do not accept, a deposit policy you never wrote.
• Copy written for a different kind of business. "Book your free consultation" on a page that sells jars of honey. It is easy to spot and easy to fix.

None of this requires expertise. It requires checking words against a life you already know. If you catch yourself thinking "I suppose that is roughly right", that feeling is a fact you have not actually verified.

Send it back so it does not come back

Sending something back is not a complaint. It is the only way the team learns the things about your business that are not written down anywhere. The difference between feedback that fixes one sentence and feedback that fixes a pattern is specificity.

• Weak: "This isn't right." Useful: "We are closed on Sundays. Never say we are open seven days."
• Weak: "Too salesy." Useful: "Drop the exclamation points and the word amazing. Write it the way I would text a regular customer."
• Weak: "The price is wrong." Useful: "The mug is $28. The $24 is last year's price and should not appear anywhere."

Three things make a correction stick. Give the right answer, not just the verdict, because "wrong" leaves the team guessing which part was wrong. Say how wide the rule goes, whether this is a fix for this one page or a rule for everything from now on. And say it to the Co-CEO, since that is the operator you talk to and the one deciding what the week contains. A rule you mention in passing about one caption tends to stay a rule about one caption.

If the same mistake comes back, do not simply correct it again. Say plainly that it is a repeat, and state it as a rule rather than a one-time fix. It also helps to keep your own short list of the facts that must never be wrong, so you are checking against something written down instead of against your memory at the end of a long day.

Fit the review into the week

The team works in a weekly rhythm. It plans, it does the work, and it reports what happened. That gives you two natural places to spend attention, and they are not worth the same.

The plan is the cheap place to steer. Reading the week's plan and saying "not that, this instead" takes a few minutes and changes what gets made. Correcting the same misunderstanding later, after five pieces have been built around it, takes five rounds. If you only have the patience for one careful read a week, spend it on the plan.

For the work itself, pick a slot and keep it. Two short sittings a week beats waiting for a calm hour that never arrives. Review in one pass rather than as things trickle in, because sorting eight items into a fast pile and a slow pile is quicker than starting cold eight separate times.

When you are away or buried, do not hold everything. Approve the reversible pieces, hold anything with a price, a promise, or a customer's name in it, and tell the Co-CEO you are short on time so the week can be planned smaller. A week with three approved things is a week that moved. A week with eleven perfect drafts waiting on you is not.

The first week costs $1, which makes it a good week to practice this rather than a week to get everything perfect. Approve some things, send some things back, and pay attention to which ones made you nervous. That is how you find out where your standards actually are.

Conclusion

Being the approver does not mean becoming an expert in web copy. It means being the only person who knows what is true about your business, and saying so quickly. The nervous version of this job reads every word hunting for anything that might be wrong. The workable version checks the handful of things that would actually cause harm, lets the rest through, and spends the energy saved on corrections specific enough that the same mistake does not return.

Before the next batch reaches you, write down the facts that must never be wrong: your prices, exactly what you sell, what you promise about delivery and returns, how someone reaches you, and anything you would need a document to prove. Keep that list next to you when you review. Then read the next piece against it, send back what fails, and approve what does not.