Choosing an evergreen and timely content mix for a seasonal event brand

The commissioning meeting runs for forty minutes and thirty of them go on the keynote recap. Who writes it, when it lands, whether the quotes need approval, whether it can be out by Friday. Nobody asks how many people will read it in November.

An evergreen and timely content mix is the decision that meeting keeps failing to make. Both kinds of piece are worth publishing. What separates a show brand with a standing audience from one that starts again every January is knowing, before the commission is placed, which of the two you are buying and in what proportion.

How do you set the ratio between evergreen and timely commissions?

Write the ratio down before the year starts, then measure against it. Sixty per cent evergreen to forty per cent timely is where I would start for a single show brand, and the defence of that number is what it does to the traffic you have to re-earn each month.

The ratio is a planning instrument, so it needs a definition tight enough that two editors classify the same commission the same way. The test I use: would this piece still be worth reading, unedited, eighteen months from now by somebody who has never heard of this year's edition? A guide to how exhibitor space is allocated passes. A profile of a first-time exhibitor passes if the profile is about the category and fails if it is about the stand. Show daily coverage fails, and should, because it is doing a different job.

Classification at commission is the part teams skip. Doing it afterwards turns into an argument about what the piece became, and the point of the ratio is to constrain what gets ordered. Where the pieces then sit in the year is a sequencing question that the year round content calendar has to answer, and a calendar built without a stated mix will drift towards timely every time, because timely pieces have obvious dates and evergreen ones do not.

Score every commission on decay half life

Half life is the useful unit here because it is the thing you are actually buying. Take the daily organic session rate a piece reaches in its first fortnight, then find how long it takes that daily rate to fall by half.

Three bands are enough for a commissioning form. Under 30 days covers recaps, announcements, award results and anything with a date in the headline. Between 30 days and a year covers agenda analysis, category trend pieces and interviews with people who will still hold that job next season. Over a year covers explainers, methodology pieces, glossaries and the how-does-this-work material that a first-time attendee searches for in any month.

A show brand that publishes nothing in the third band has a site with no floor. Every session it earns in June is a session that some piece published in May had to go and win.

Predicting the band at commission is less mystical than it sounds, and four questions get you most of the way. Does the headline contain a year, an edition or a date. Does the argument depend on one named person holding one named job. Would a reader searching this subject in a month with no show running find the piece useful. Is the search demand for the subject roughly flat across the year, which you can read off any keyword tool in a minute. Three noes and a yes to the last one puts the piece in the long band with reasonable confidence.

What the mix buys you, in arithmetic

Take a desk publishing 120 pieces a year and work the two ratios through with illustrative figures.

Assume an evergreen piece settles at 25 organic sessions a month from its third month onward and holds there. Assume a timely piece takes 900 sessions in its first month and 15 a month after that.

At 60 to 40, that is 72 evergreen and 48 timely. Once the evergreen back catalogue has matured, it contributes 72 times 25, which is 1,800 sessions a month with nobody doing anything. The timely side publishes four pieces a month, contributing 3,600 in-month sessions, and the 44 older timely pieces contribute 44 times 15, which is 660. Monthly total 6,060, of which 3,600 has to be won again next month.

At 30 to 70, that is 36 evergreen and 84 timely. The evergreen base drops to 900. The timely side publishes seven a month for 6,300 in-month sessions, with a tail of 77 times 15, or 1,155. Monthly total 8,355, which is 38 per cent higher, and 6,300 of it, 75 per cent, is re-earned monthly.

Now stop publishing for two months, which is what happens to most show desks after the edition closes. The 60 to 40 site falls to roughly 2,460 a month. The 30 to 70 site falls to roughly 2,055 despite having been the bigger site. The higher-output configuration bought volume that evaporates exactly when the desk is least able to replace it, and the publishing cadence between editions is the other half of that same failure.

Run this with your own numbers and the crossover point moves. That is the argument for running it rather than adopting my 60.

The cost side pulls the other way and deserves saying plainly. An evergreen explainer takes longer to commission, needs a writer who understands the mechanics, and usually needs a subject expert to look at it before it goes out. Call it a day and a half against half a day for a recap. It also takes two or three months to reach the base rate the arithmetic above assumes, so a desk that switches to 60 per cent evergreen in January sees higher costs immediately and the return in the second half of the year. That lag is why the ratio has to be a stated policy with a named owner. Anything decided quarter by quarter reverts to recaps by March.

Why does search treat fresh and evergreen pages differently?

Google's ranking systems guide, last updated in December 2025, lists the systems it will name publicly, and two of them decide this question. On freshness it says: "We have various 'query deserves freshness' systems designed to show fresher content for queries where it would be expected." On originality it says Google has systems "to help ensure we are showing original content prominently in search results, including original reporting, ahead of those who merely cite it."

The conditional in the first quote is the whole point. Recency helps where recency is expected. A search for a show's 2026 agenda expects this year's page. A search for how badge scanning works at a trade show expects nothing of the kind, and a well-made 2023 explainer will hold that position against anything you publish this week.

The second quote cuts the other way and is the reason a timely piece can still be a good buy. Original reporting from the floor of your own show is content nobody else can produce, and Google says it has systems intended to rank it ahead of the sites that pick it up afterwards. A recap that just lists what was said has no such protection.

Checking the prediction at day 90

Predicted half life is a guess until you check it, and the check is cheap.

Pull every piece published between 120 and 90 days ago. For each, take organic sessions in days 1 to 30 and in days 61 to 90, and compute the ratio of the second to the first. Anything above 0.7 behaved as evergreen. Anything below 0.15 behaved as timely. The middle band is where your commissioning form was wrong, and reading twenty of those pieces will tell you more about your desk than any keyword report.

Then compare the realised split against the stated ratio. A desk that commissioned 60 per cent evergreen and produced 35 per cent evergreen behaviour is not lying to itself deliberately. It has an idea of what evergreen means that the audience does not share.

The decay curve may declare itself earlier than 90 days. Castillo, El-Haddad, Pfeffer and Stempeck, in their 2013 work on the life cycle of online news stories, modelled the total traffic an article would ultimately receive by observing the first ten to twenty minutes of social media reactions, and reported improvements in the early prediction of shelf life. Their setting is a news publisher with heavy social distribution, so do not import the exact method into a trade brand where most traffic arrives from search over months. The transferable point is that shelf life is estimable early, which makes the day 90 check a validation exercise instead of an autopsy.

Where this stops

Half life is only knowable in advance for the boring cases, and the boring cases are the ones you did not need help with. Everybody can tell that a keynote recap decays and a glossary does not. The commissions that decide whether a mix is any good sit in the middle band, where a piece on a regulatory change is evergreen for two years and then worthless overnight because the regulation moved.

Seasonality also confounds the day 90 measurement badly for event brands. A piece published six weeks before your show sits in a rising demand curve and one published two weeks after sits in a collapsing one, so the same article can measure as evergreen or timely depending on when it went out. The correction is to compare pieces published in the same month against each other and never across the show boundary, which means your first honest read takes a full annual cycle.

Nothing here says what to do with the pieces that decayed and stayed decayed. Some of them still earn a trickle, some are actively misleading about an edition that has passed, and the disposal rules for those sit with archiving past edition content and with the wider content studio question of what a portfolio keeps.

Take the last 30 pieces your desk published, tag each one evergreen or timely from the headline alone, and count. If the split is a long way from whatever ratio you would have said you were running, you have found the problem before spending a day in the analytics.