Year-End Website Planning: Your Fix, Keep and Build List for 2027
20 November 2026 / 5 min read

Most businesses treat their website like a house they live in: they notice what is broken when it starts to leak.
The last quarter of the year is the best time to do something more deliberate. Budgets are being set, teams have a clear view of what worked, and the busiest season has either arrived or is about to. Google's year has also given you plenty to react to: two core updates, four spam updates, the rise of AI answers and the retirement of FAQ rich results.
This is a planning framework, not a checklist. If you want the detailed technical review, our 40-point website audit checklist is the companion piece. Here the question is different: out of everything you could do, what should you do next?
Step one: look back honestly
Before planning anything, spend an hour with the evidence. You need four things:
Enquiries by month, and where each came from.
Your ten most visited pages and the ten most valuable ones, which are rarely the same list.
Search Console data for the past six months, with the dates of Google's updates marked.
What your sales team or you hear, such as the questions prospects ask, the objections that recur and the pages they say they read.
Then answer three questions in writing.
What brought us business this year?
What did we expect to work that did not?
What did we never measure?
The third usually produces the most useful work for the new year.
Step two: sort everything into four buckets
Gather every idea, complaint and suggestion about the website into one list, then sort each item.
Bucket | What goes in it | Examples |
Fix | Things that are broken or costing you now | Slow mobile pages, inconsistent contact details, duplicate pages, conflicting structured data, empty image descriptions, a back button that misbehaves |
Keep | Things that work and should be left alone | Your best-performing pages, a case study that wins meetings, a form with a good conversion rate |
Retire | Things that no longer earn their place | Thin pages, outdated posts, unused tags, plugins and scripts nobody owns, FAQ markup you maintain for a retired feature |
Build | New things worth creating | A page for a service you now sell, a case study from this year's best project, an original data piece, a directory profile |
The "keep" bucket is the one people forget, and it prevents the most expensive mistake: redesigning something that was working.
Step three: score by impact and effort
For each item in Fix and Build, give two scores from one to five: how much it would help, and how much work it is. Then plan in this order:
High impact, low effort. Do these first. Fixing mobile speed or contact-detail inconsistencies often sits here.
High impact, high effort. Schedule these and give them an owner and a date.
Low impact, low effort. Batch them into one afternoon.
Low impact, high effort. Cancel them. This is where most vanity projects live.
The discipline is in the last category. A list of thirty tasks, all given equal weight, is how nothing gets done.
Step four: decide whether you need a fix or a redesign
Some years the honest answer is a redesign. Most years it is not.
A redesign makes sense when the structure is wrong for your business, the platform limits what you need to do, or the problems are structural rather than fixable. It does not make sense when the real problems are thin content, missing proof or an unclear call to action, because those travel with you into the new design.
If you suspect a redesign is due, read our step-by-step account of the redesign process before you start, particularly the sections on content readiness and redirects.
Step five: respect the calendar
The last quarter has constraints that the rest of the year does not.
Do not launch major changes just before your peak season. If your business depends on the festive and year-end weeks, freeze structural changes before then. A new design that needs two weeks of bug fixing is a poor way to enter your busiest period.
Do the quiet work now. Content, measurement setup, photography, directory profiles and the technical fixes that do not change what visitors see are all safe to do in busy periods.
Build the January plan in December. Agencies and freelancers fill up in the new year. Agreeing scope and priorities before the break means work can start the first week back.
What your 2027 website plan should include
A workable plan is short. One page covers:
Section | Contents |
Goal | One number the website exists to move, such as qualified enquiries per month |
Evidence | The three findings from your look back that drive the plan |
Fix list | Up to five items, each with an owner and a date |
Build list | Up to three items, each with an owner and a date |
Measures | The scorecard you will review monthly |
Review dates | A short quarterly check-in, with the Google update calendar |
If it does not fit on a page, it is not a plan yet.
A 90-day timeline
Period | Focus |
November | Look back, build the Fix, Keep, Retire and Build list, and finish any high-impact, low-effort fixes. Avoid structural changes during peak weeks. |
December | Agree the 2027 plan, scope the larger items and book the work. Publish the content that has already been prepared. |
January | Start the first high-impact project. Set up the monthly scorecard. Review the first month's results. |
Questions to put to your agency or team
What are the three things on our site most likely to be costing us enquiries?
Which pages should we leave alone, and why?
What did this year's Google changes mean for us specifically, rather than for the industry?
What would you stop doing if we asked you to spend half as much time?
How will we know in March whether this plan is working?
A good partner will answer these without defensiveness. A weak one will change the subject to a redesign.
If you would like the leaks found for you, our post on the conversion leaks we find on almost every site is a good starting point, and our B2B website checklist covers what enterprise buyers look for before they enquire.
Frequently asked questions
When is the best time to review a website? Once a year at minimum, and the last quarter is ideal because you have a full year of data and can plan before budgets are spent.
How do I decide what to fix first? Score each issue by impact and effort. Start with high-impact, low-effort fixes, and cancel low-impact, high-effort ones.
Should I redesign my website for 2027? Only if the structure or platform is the problem. If the weaknesses are content, proof or calls to action, you can fix those on your current site.
Should I change my website before the festive season? Avoid launching structural changes just before peak demand. Quiet improvements, such as content, tracking and speed fixes, are generally safe.
How long should a website plan be? One page. If it needs more, the priorities are not clear enough.
Want an independent view before you plan? Our free audit gives you a ranked list of what to fix first, in a Conversion Leak Report you can take straight into your planning. Request your Free Website Audit →
Working on something like this? We do web design and development, and the first call is free.

