A trending article surges to 20,000 reads in three days, drops to single digits by day seven; an evergreen article published the same period only gets 300 reads in the first week, but six months later steadily brings 200 organic searches per day. Put both curves on one chart, and the answer to how to arrange the ratio becomes clear.
Figure: Key points of evergreen vs trending content: how to arrange the ratio without flipping (compiled by 运营GO)
Quick conclusion: the reasonable starting point for most small-to-medium teams is evergreen 7, trending 3, with quarterly fine-tuning using real data. The value of trending isn’t in long-term traffic, it’s in new-user exposure and topic占位; the value of evergreen is in compound interest still going six months later. Judging the type wrong is more fatal than an imbalanced ratio — four judgment questions can定性 a topic in 3 minutes at a topic meeting.
Where the traffic curves of the two content types differ
Draw the same batch of articles by daily average visits for 180 days after publishing, and the shapes are clearly separated. Trending articles are spike-type — high peak, fast decay, eating 80% of full-lifecycle traffic within seven days. Evergreen articles are climb-type — almost no volume in the first 30 days, entering search ranking climb period from day 60 to 90, then entering a plateau.
| Dimension | Evergreen content | Trending content |
|---|---|---|
| Traffic peak | Low, accumulates over time | High, concentrated in 3-7 days |
| Effect cycle | 60-120 days | Within 24 hours after publishing |
| 180-day retained traffic | Over 70% of full cycle | Less than 10% |
| Main source | Organic search | Social media recommendations and shares |
| Per-article maintenance cost | 1-2 updates per year | Basically no maintenance needed |
| Suitable goals to handle | Lead accumulation, brand search | Exposure & new users, topic participation |
The most easily overlooked row in this table is the last one. Using trending articles to assess lead conversion, or using evergreen articles to assess first-week exposure, are both metric mismatches. Assessment criteria should follow content type — the specific breakdown can apply the layered attribution method in the content ROI data model.
Start at 7:3, then shift both sides with data
7:3 isn’t the standard answer, it’s the lowest-risk starting point. The faster your industry changes, the higher the trending ratio can be; the longer the product decision cycle, the more the evergreen ratio should be increased.
| Team situation | Evergreen ratio | Trending ratio | Judgment basis |
|---|---|---|---|
| New site cold start, low domain authority | 80% | 20% | Search rankings need time, accumulate assets first |
| Already has stable organic traffic | 70% | 30% | Evergreen as base, trending for incremental |
| FMCG or policy-sensitive industry | 55% | 45% | Topic window short, absence equals silence |
| B2B high ticket, long decision chain | 85% | 15% | Buyers repeatedly search similar questions |
- Ratio is calculated by article count, not word count — trending articles are usually shorter
- Keep at least 4 evergreen articles per month — below this number, a traffic gap will appear six months later
- Don’t let trending articles占用 main writers — hand them to fast-reacting freelancers or the editor themselves
Four questions to judge a topic’s true type
The most common dispute at topic meetings is “does this count as evergreen.” Four questions can定性 it — only all “yes” counts as evergreen.
- One year from now, if you search this keyword again, does the reader’s need still hold?
- Does the core conclusion in the article not depend on a specific date, a certain press conference, or a certain new regulation?
- Is the 12-month trend line for this keyword in search tools steady, without a single peak?
- Among similar competitor articles, are the top 3 publish dates all over one year old?
The third question has the most discriminatory power. Keywords whose trend line shows a single spike — no matter how comprehensively you write it, nobody will search it six months later. This type of keyword should be done, but scheduled and assessed as trending.
There’s also a mixed type: the topic is evergreen but the cases carry timeliness, such as annual ranking lists. This type is classified as evergreen, but must register an annual update task in advance — the update rhythm references the experimental conclusions of the content update frequency experiment, don’t wait until it drops out of the top ten pages to remember.
Leave a jump-the-queue channel for trending content
The biggest enemy of trending is rigid scheduling. By the time you go through the three processes of topic meeting, brief, and review, the window has already closed. The feasible approach is to reserve fixed capacity, not temporarily grab resources.
- Every week,固定 leave 1.5 article slots without specific topics,专门 for trending
- Trending articles go through a simplified process: verbal alignment replaces written brief, editor directly does final review, skip secondary review
- Set a hard line: trending that can’t be published within 72 hours is directly abandoned, converted to an evergreen angle rewrite
- Do A/B testing separately for trending article titles — this type of content is more sensitive to click-through rate
Put evergreen scheduling and trending channels into the same calendar, use the dual-axis view to distinguish main lines and jump-the-queue lines — the operation method can directly reuse the dual-axis content calendar scheduling method. Use colors on the calendar to distinguish the two types of articles — when the ratio is imbalanced, you can see it at a glance.
Calculate evergreen content maintenance cost in advance
Evergreen doesn’t mean once-and-for-all. A tutorial that’s been live for two years usually needs to go through 2-3 updates to hold its ranking. When team scale reaches 100 evergreen articles in stock, maintenance itself will eat up 20% of capacity.
- Scan stock once per quarter, pick out articles whose rankings dropped more than 5 positions and queue them for update
- Update priority = former monthly average traffic × decline幅度, not scheduled by publish time
- Maintenance capacity gets a separate budget, don’t mix it with new article capacity in one pool competing for people
To judge whether an article has entered the decay period, look at the three signals given by the content decay curve: stable impressions but declining clicks, decreasing long-tail keyword count, average ranking retreating for two consecutive months. If two of the three signals appear, it’s time to act.
Review once every three months, see which side to shift the ratio
Review only looks at two numbers: the 90-day retained traffic ratio of evergreen content, and the first-week exposure contribution ratio of trending content.
- Evergreen retained ratio below 50%: means topics are leaning trending, tighten the four judgment questions
- Trending first-week exposure contribution below 15%: means trending isn’t hitting the mark, reduce ratio or change channels
- Both healthy but total volume not growing: the problem is in distribution and internal links, not in ratio
There are four actions you can take this week: export the 180-day traffic curves of all articles from the last 6 months, split into two piles by spike-type and climb-type, calculate your current true ratio; make the four judgment questions a fixed环节 of topic meetings, mark the type for each topic on the spot; in next month’s calendar, leave 1.5 trending article slots per week, write down the simplified process; build a maintenance queue table for existing evergreen content, sort by traffic × decline幅度, handle the top five first.
FAQ
What ratio of evergreen to trending content is stable?
Most small-to-medium teams start at evergreen 7, trending 3, then quarterly fine-tune with real data. New site cold start goes to 8:2, FMCG industry can go to 5.5:4.5.
How to judge whether a topic is evergreen or trending?
All four questions “yes” counts as evergreen: need still exists one year later, conclusion doesn’t depend on specific date, trend line steady without single peak, competitor top 3 publish times over one year.
Should trending content be done at all?
Yes, but assess it as trending. Its value is in new-user exposure and topic占位, not in long-term traffic; it eats 80% of full-lifecycle traffic within seven days, don’t use it to assess lead conversion.
Is evergreen content done once and left alone?
No. A tutorial live for two years usually needs 2-3 updates, when there are 100 evergreen articles in stock maintenance eats 20% of capacity, update priority sorted by traffic × decline幅度.
What if trending topics always miss the window?
Every week固定 leave 1.5 article slots for trending, go through simplified process, if can’t publish in 72 hours abandon it, convert to evergreen angle rewrite — don’t let scheduling kill the topic.
