Approval management

What approval management is

Approval management is the chasing. I take each finished piece to the person who has to sign it off, give them a date, tell them what happens if the date passes, remind them before it does, and record the decision in writing when it comes. It is not a clever system and it is not software. It is largely polite persistence, done by somebody from outside the business who is not worried about asking a director for the fourth time.

It is worth paying for because sign-off is where most content quietly dies. Not dramatically, and not with an argument. A draft goes to somebody who means to read it that week, then the quarter-end arrives, then it is three weeks old and nobody wants to raise it, and eventually it is out of date and the whole thing was wasted. Almost every business with a drawer full of nearly-finished content has this one problem and no other.

The design half is a separate job. Deciding who approves what, in what order, what a deputy is allowed to do and what the default should be is approval workflow, in the strategy pillar. That is the rule book. This is somebody enforcing it week after week, which turns out to be the harder of the two.

When you need one

The plainest sign is drafts you cannot age. If you cannot say how long the oldest unapproved piece has been waiting, it has been waiting too long. The second sign is a single bottleneck approver, usually the owner or a technical specialist, who is genuinely busy and genuinely necessary and receives every draft personally.

The third is regulated or technical content, where a claim really does need checking by a qualified person before it goes out. That check is legitimate. What is not legitimate is it taking eleven weeks because nobody owns the reminder.

You do not need this if you approve your own work. One person writing and publishing has no approval stage, and adding one invents a delay that was not there. If you have staff and no agreed order at all, do not buy chasing yet; buy the sign-off path first, because chasing without an agreed default is just nagging.

What I actually do

• Send one version, clearly named. Approvers are given the current draft and told it is the current draft, so nobody comments on last week's file. This runs on the same rules as revision management.
• Ask for a decision, not an opinion. Each piece goes out with three options: approve, approve with the changes listed, or reject with a reason. Open-ended requests for thoughts are how a draft acquires four rounds of comments and no decision.
• Set the date and say the default. Five working days is typical. The default is agreed in advance and written on the request: the deputy decides, or the piece publishes with the approver's changes made and a note recording that it went unreviewed.
• Remind, twice, in different ways. Once at the halfway point and once the day before. The second reminder goes wherever that person actually answers, which is not always email.
• Record it. Who approved what, on what date, on which version. Filed with the piece. This is the part that matters six months later when somebody asks who agreed to a claim.
• Report the blockage. Where a piece is genuinely stuck, you get told which named person is holding which piece and for how long, in a list, once a week. Most of them unstick themselves at that point.

What you get

Approvals, in writing. The actual decisions, dated and attached to the version they apply to, so the piece can be published without anyone reopening the question.

A weekly held list. One short list: piece, approver, days waiting. No commentary. It is usually enough on its own.

A record of defaults used. Where a deadline passed and a piece moved anyway, that is written down, with the date and the rule it was moved under. Nothing happens quietly.

A worked example

An illustration, not a client. A twelve-person physiotherapy clinic with three locations wants to publish nine condition pages and a fortnightly newsletter. Every page has to be read by the clinical director, who treats patients four days a week and is the right person to check the claims. Eight months in, two pages are live and seven drafts are sitting in an inbox.

The change is not a new tool. Each page goes to the clinical director as one named file with three boxes to tick, on a Monday, due the following Monday, with a written default: unreturned pages go to a second senior physiotherapist. The first week two pages come back approved and one is returned with corrections. The default is used once, in week four, and the clinical director objects to it, which is the point at which the remaining pages start coming back inside three days. All nine are live within seven weeks, and the newsletter, which needs no clinical reading, stops waiting behind them.

How it runs

1. A half-hour call. What is stuck, who has to sign it, and how long it has been sitting. No charge.
2. A fixed price in writing. Agreed before anything starts, by the month or by the project.
3. The approver list. Names, what each one signs, their deputy, and the default when a deadline passes. Written on one page and agreed with you, not assumed.
4. The first round. Everything currently waiting goes out at once with dates on it, in the first week. The backlog is usually where the fastest result is.
5. The weekly rhythm. Requests out on the same day each week, reminders midway and the day before, held list to you at the end of it.

What it costs

Quoted in CAD after the half-hour call, usually monthly, because chasing is ongoing work rather than a deliverable. The price depends on how many pieces a month need sign-off and how many approvers are involved. For most businesses this is not bought on its own; it arrives inside content production management, where it is one stage of the pipeline. The pricing page shows the four ways I work, and whichever applies, the number is fixed in writing before anything starts.

What happens next

Once approvals move, the next stage usually exposes itself, and it is often version confusion, which one current version fixes. If the approvers keep asking the same question about claims and permissions, the answer is a pre-publication review before the piece reaches them rather than after. And if the same argument about who decides recurs, it belongs in content governance, written once. None of that follows automatically. Getting the sign-offs in is a complete piece of work and nothing further is assumed.