Ask someone to recall a list and they will reliably remember the beginning and the end better than the middle. This has been measured since the 1960s, holds across list lengths, and applies to almost any sequence a person processes in order.
Your navigation is a sequence. So is your service list, your process description, and the order of sections on a page. The research has a direct application, and almost nobody uses it deliberately.
In this article
The study
Bennet Murdock’s 1962 paper The Serial Position Effect of Free Recall established the shape.
Participants were shown lists of words, ranging from ten to forty items, each displayed for one second. They then had ninety seconds to recall as many as they could, in any order.
Plotted by position, recall formed a consistent curve. Items at the start were recalled well, an effect extending over roughly the first three or four items. Items at the end were recalled well, extending over roughly the last eight. Between those, a long flat stretch where recall was poor.
The result held regardless of list length. Longer lists did not mean a longer memorable section. They meant a longer forgettable middle.
Why it happens
The two effects have different causes, which matters for applying them.
Primacy comes from processing. The first items arrive when attention is fresh and nothing is competing. They get more rehearsal and are more likely to be stored durably.
Recency comes from availability. The last items are still immediately accessible when recall begins. This effect is fragile. Introduce a delay or an intervening task and recency weakens considerably, while primacy largely survives.
For a website that difference is practical. If someone is deciding while looking at the page, both positions are working. If they are recalling your company in a meeting three days later, primacy is doing most of the work and the last item has largely faded.
Recall of a list forms a curve: strong across the first three or four items, strong across the last eight, and weak through everything in between.
Murdock, The Serial Position Effect of Free Recall, 1962
Where this applies on a B2B site
Navigation order. Most B2B menus are ordered by internal convention, running Home, About, Services, Industries, Case Studies, Blog, Contact. That order reflects an org chart, not a priority. The first and last items get remembered. If Contact is last, that is at least deliberate. If your highest-value page is fourth of seven, it is in the forgettable middle.
Service lists. As covered in how choice research explains shorter service pages, a list of five services is not read as five equal items. The first is remembered, the last is remembered while still on screen, and the middle three blur.
Section order on a page. The opening section and the final section carry disproportionate weight. The final section on most B2B pages is a generic call to action, which spends a strong position on a weak message.
Process descriptions. A five-step process is remembered as its first and last steps. If the step that differentiates you is step three, it is invisible.
Case study structure. The situation at the start and the outcome at the end are what survive. Method in the middle is read by the technically interested and forgotten by everyone else, which is the correct allocation as long as you know it is happening.
What to do with it
The instruction is not to reorder everything by importance. That produces pages front-loaded with claims and no structure.
The useful applications are narrower.
- Put the thing you most want remembered first, not third. This is the single highest-value application, because primacy is the durable effect.
- Do not waste the last position. The final section of a page is a strong slot. A specific next step with a stated response time uses it. “Get in touch to learn more” does not.
- Keep lists short enough that the middle is small. Three items have almost no forgettable middle. Nine items are mostly middle.
- Break long lists into groups. Grouping creates several first and last positions. A grouped list of nine, in three groups of three, has more memorable positions than a flat nine.
- Repeat what matters across positions. If one message must survive, it should appear early and be restated at the end, in different words.
- Order deliberately, and write the reason down. The value here is mostly in replacing an accidental order with a considered one.
The limit of the research
This is memory research, conducted with word lists in controlled conditions. A website is not a word list, and visitors are not being tested on recall.
Two cautions follow.
The effect describes what is remembered, not what is persuasive. A buyer may remember your first service and choose your third, because the third was the relevant one. Position influences recall, not suitability.
And a website is scanned rather than presented sequentially. Someone who reads only headings has constructed a different sequence from the one you laid out. Serial position applies to the order they processed, which is why the interaction with scanning behaviour matters. The effective list is your headings, not your paragraphs.
So this is not a large effect on a single page visit. It is a compounding one across the several encounters that make up a B2B buying process.
The middle is not wasted space
A common overcorrection is to conclude that the middle of any sequence is dead and should be stripped out.
That is wrong for two reasons.
The middle is where detail belongs. Not everything needs to be remembered. A buyer does not need to recall your quality assurance process; they need to be able to find it and satisfy themselves that it exists. Content that serves verification rather than recall is correctly placed where recall is weakest.
And the middle is what makes the beginning and end credible. A page consisting only of memorable claims, with nothing substantiating them in between, reads as marketing. The middle is where the argument is made. It is read carefully by the minority of visitors who are closest to buying, which makes it commercially important even though it is poorly remembered.
The distinction is between what should be retained and what should be available. Put the first in strong positions. Put the second where it can be found.
Across pages, not just within them
The same effect operates at the level of a whole visit.
A buyer’s session has a sequence. The page they landed on, the pages they moved through, and the page they left from. The landing page carries primacy. The exit page carries recency, at least briefly.
Most B2B sites give no thought to the exit page, and for many visitors it is a case study or a service page that ends without a next step. The last thing seen is a page that stops.
This is the argument for a consistent, specific closing element on every page that a visit could plausibly end on. Not a banner repeated for its own sake, but a genuine next step appropriate to that page. It occupies the recency position for anyone whose visit ends there, and on a multi-page site that is most visitors on most pages.
Why it compounds in B2B
The reason to care about recall in B2B is that the decision is not made while anyone is on your website.
It is made later, in an internal conversation, by people working from memory and a few notes. Someone says “who else did we look at”. Someone else answers from what they retained.
At that moment, what matters is not how thorough your site was. It is what survived. The company remembered as “the one that does X” is in the conversation. The company remembered as “there was another one, I forget what they specialised in” is not.
Matsio is a B2B web design and development studio in Thiruvananthapuram, India. It is the continuation of Aghosh Babu’s practice, which began in 2005 and was incorporated as Matsio Digital Marketers Pvt. Ltd. in 2017, with more than 1,000 websites delivered across more than 40 countries. The studio’s own positioning is stated in the same words in the same position across every page, because consistency of the first thing said is what makes it repeatable by someone else later. 85% of clients return for further work, measured across every engagement since 2005.
An exercise
Write down, in one short phrase, the thing you want a buyer to say about your company in a meeting you are not in.
Then open your homepage and find where that phrase appears. If it is not in the first section, and not in the last, it is in the middle of your own page, competing with everything else for a position the research says is weak.
Then do the same for your navigation, your services list and your process. In each case, check what is first and what is last, and whether either was chosen on purpose.
Most teams find that their strongest differentiator sits somewhere in the middle of three different sequences, put there by the order it was written in rather than by any decision.
Related reading. social proof research covers what makes a claim repeatable once it is remembered, and how answer engines choose which companies to mention covers the same recall problem when the thing doing the remembering is a machine.
The short version
People remember the start and the end of a sequence and lose the middle. Primacy is the durable effect; recency fades once attention moves on.
Your navigation, service list, process and page sections are all sequences. Put what you most want remembered first, use the final position for something specific, and keep lists short enough that the middle is small.
A small thing and a big thing
One small thing to fix on your website today, and one big thing to learn that gets you more leads.
Your information is
safe with us. No spam.