Between the 3rd and the 10th of August 2026, across six retained accounts, we closed forty-six pieces of work.
Fourteen were HTML award mailers for a luxury travel awards programme, one per shortlisted property, and all fourteen went out in the same week. Each hand-built. Each carrying different photography, a different property, a different set of names that had to be spelled correctly, and the same programme identity holding all fourteen together so they read as one family rather than fourteen separate emails.
Eighteen were for a single hospitality account. A monthly social media performance report and then the July one. An October product feature. A home page banner. A specials update. A menu piece. A guesthouse section added to the header navigation. A signup form that had stopped accepting signups. A page that had started loading slowly. A subscription that failed and needed chasing.
Five were display faults on another retained site, all found and fixed in one afternoon. Testimonials not showing. A community page not rendering. Images missing from a section that needed them.
Three were for a publisher: a video embedded on a page, a change to the template behind a recurring editorial feature, and a round of theme and plugin updates.
Two were for a health education organisation, a newsletter performance report and a standing catch-up.
Two were for a retail account, an analytics insight that arrived unprompted and a product feature scoped off the back of it.
And twice in the same week, community management on our own channels. It gets closed, reopened and closed again, because it is a task that is never actually finished.
None of it will ever be a portfolio piece.
Forty-six is the wrong number
Forty-six is also too low, and the reason why is worth more than the count.
On some accounts we close almost nothing. Not through sloppiness. The work simply does not end, it recurs, and marking it done would be a small lie. A task gets opened, delivered, handed back, and then four months later the same task wakes up with a fresh brief attached to the bottom of it.
One of ours was opened in February 2024 for a set of company brochures. It is still open this morning. Somebody marked it complete once, in September 2024, and reopened it four days later. The headline statistic printed inside those brochures has been revised four separate times since, because the company kept growing, and every revision means going back through each brochure and each regional variant to find every place the old number appears. Four of them were delivered again on the 4th of August, inside the week described above. The task is still open, because it will be back.
Another is one image. A single statistics panel for a landing page, opened last August, delivered three times. Each time the numbers had moved.
A third was a conference stand, briefed as copy last year’s. Halfway through it stopped being that and became a timeline design listing every customer and the year they arrived. More than twenty comments before sign-off. Also still open.
None of that appears in the forty-six. It is the same work, done in the same week, invisible to the count for reasons that are purely administrative.
None of this is a project
Read the list again and notice what is missing. There is no launch in it, no rebrand, nothing with a kick-off deck and a go-live date. Nothing that would survive being written up as a case study, because a case study needs a before and an after, and none of these have either.
What each one is instead is a small decision about how a brand shows up that week.
Which photograph leads the mailer. Whether the October feature sits above or below the seasonal banner. Whether the newsletter report says the open rate went up, or says the open rate went up because the send moved to a Tuesday. Whether the editorial template still matches the site after the site changed. Whether the statistic on the opening spread of the brochure still matches the one buried near the back. None of these are hard problems. All of them are decisions, and all of them get made whether or not anybody qualified is making them.
The work is invisible precisely because it is being done.
What drift looks like
Brand drift never announces itself. Nobody sends an email to say the brand has started coming apart. It arrives instead as a run of individually reasonable choices.
The logo goes out at the wrong weight because the file closest to hand was the wrong file. A template stops matching the site, because the site moved and the template did not. A seasonal push gets designed by whoever had a free afternoon, in whichever tool they knew, and it looks fine on its own and slightly wrong next to everything else. Somebody orders branded stickers and they come back teal, because the hex value was never written down anywhere a printer could reach it.
Six months of that and the brand has not been damaged by any single decision. It has been diluted by forty of them.
Nothing on the list above prevents a disaster. Collectively they prevent six months of small, reasonable, uncoordinated choices.
Nobody owns the week
The common alternative to a design retainer is not doing nothing. It is assembling the same capability out of parts: a strategist for the thinking, a freelance designer for the artwork, a developer for the site.
That arrangement can work, and for some businesses it is the right shape. Where it leaks is at the joins, and specifically on the ordinary weeks.
The strategist is engaged when there is strategy to do. The designer is briefed when there is a brief. The developer is called when something breaks. All three can be excellent and none of them is watching the week. So the October feature gets designed without anybody checking what November will need. The mailer template drifts away from the site template, because two different people own them. Nobody remembers that the brochure statistic changed in March, so the version going to the printer in August is still carrying the old one. And the question of whether all of it still looks like the same company falls to the person who commissioned all three, who is usually the owner, and who did not hire three specialists in order to become the fourth.
The week has to be somebody’s job. On a retainer it is ours.
What you are actually buying
Hours are how a design retainer gets measured. They are not what it is for.
What it buys is continuity of judgement. The difference between a team that has to be told the campaign date and a team that already knows it. Between briefing the brand every time and briefing only the work. Between explaining, again, why the secondary typeface is not the display typeface, and simply having that be understood.
It buys memory, too, which is the part nobody puts in a proposal. Somebody has to know that this brochure exists in five variants, that two of them are Australian, and that the low-resolution versions are the ones that go to the conference app. That knowledge is not written down anywhere. It lives in whoever has been doing the work.
And it buys speed, though not in the way people expect. A retained team is not faster at making an email. It is faster at deciding what the email should be, because the last eleven decisions are already in the room.
Look at the fourteen mailers again. Fourteen properties, one identity, one day. That is not a difficult brief. It is a brief that only works if the same team holds all fourteen at once and remembers what the other thirteen looked like.
What a design retainer is not
Two things worth being plain about, because the vagueness helps nobody.
It is not an unlimited-work bucket. A retainer is a committed volume of attention and it has a shape. When something large turns up in the middle of it, a full rebuild, a new site, a rebrand, that is a project, and it gets scoped and quoted as one. Pretending otherwise produces a retainer that quietly eats itself, and an agency that starts resenting the client, which is a worse outcome for everybody than an honest conversation in month one.
It is also not a discount. It is not the cheap way to buy design work by the hour. If all you need is artwork produced at volume then there are cheaper ways to buy artwork, and we would tell you so. The value sits in the continuity, and if you do not need the continuity you are paying for something you will not use.
Where to start
The useful conversation about a design retainer is not about price. It is about scope. What actually happens in your ordinary weeks, who is currently making those decisions, and whether the answer is anyone in particular.
If you can list your last month the way we have just listed ours, and the list runs longer than you expected and shorter on owners than you would like, that is the conversation worth having. Tell us what your week looks like and we will tell you honestly whether it needs one.
