Web Design
7 Website Redesign Checklist Items: Complete 2026 Guide

Joshua Jackai
16 min

Published September 2026 ยท Last updated and fact-checked 3 November 2026
Short answer: A website redesign checklist lists what a business needs ready before design begins. I use seven items: a goal with a baseline metric, a content inventory, brand assets, account access, one named decision-maker, a launch date with buffer, and a redirect and tracking plan. Published checklists run from 6 steps (Delta) to 23 items (Buarich).
Key takeaways
Write the goal as one sentence with one metric, then record what the site produces today. That number is the only thing the new site can be judged against.
Nidumolu (1995) studied 64 software projects and found that coordination run by authorized entities reduced project uncertainty. A website is smaller, so I use it as direction for item 5: one named approver.
In a laboratory study of rated screenshots, people formed visual-appeal judgments within 50 milliseconds (Lindgaard et al., 2006), so brand assets and first-screen proof go in the kit, not in the last week.
Late copy is the most common cause of a slipped redesign in my projects. Name who writes it and the date it lands.
Redirects and a tested form submission belong on the checklist before design starts, because both are cheap on paper and expensive after launch.
# | Checklist item | Who prepares it | Time to gather | What slips if it is missing |
|---|---|---|---|---|
1 | Goal and baseline metric | Marketing lead | 1 to 2 days | The redesign is judged on taste |
2 | Content inventory | Marketing lead and writer | 3 to 5 days | Pages are lost or rebuilt twice |
3 | Brand assets | Marketing lead | 2 to 3 days | Design starts from a screenshot |
4 | Access and logins | Whoever owns each account | 1 to 2 days | Build waits on a password |
5 | Decision-maker and reply time | Executive sponsor | 1 day | Review rounds stretch |
6 | Launch date and buffer | Project owner | 1 day | The date has no buffer |
7 | Redirects and tracking plan | Designer and analytics owner | 2 to 3 days | Traffic and leads vanish at launch |
Times to gather are my estimates from client work, not a study.

1. Write the goal and the baseline metric: the item that settles every later argument
A redesign needs one primary goal written as a sentence and one metric that proves it. Record today's value for that metric before anything changes, so improvement is a measurement and not an opinion. Rank two or three secondary goals so trade-offs settle themselves.
Owner: marketing lead, with sales for the lead definition
Time: 1 to 2 days
Output: one sentence, one metric, one baseline number, one target date
Common metrics: form fills, booked calls, qualified leads, pipeline created
In my experience teams skip the baseline because the old numbers are uncomfortable. That is exactly why to record them. In my projects the redesigns that got judged fairly all had a number written down before design began, and the ones that got argued about did not.
Pick the metric a sales team would recognise. A qualified lead beats a visit, because a visit can rise while pipeline falls. Pull twelve weeks of data so a single odd month does not set the baseline, and note any campaign, seasonal peak or one-off promotion that inflated it. Write the target date next to the target number, because a goal with no date can never be missed and so never gets reviewed.
Kohavi, Longbotham, Sommerfield and Henne wrote a survey and practical guide on running controlled experiments on live websites (Data Mining and Knowledge Discovery, 2009). Testing needs a stated metric, and a redesign with no baseline has nothing to compare against. A redesign without a stated metric gets judged on taste, and taste changes with every new reviewer.
"A redesign without a stated metric gets judged on taste, and taste changes with every new reviewer," wrote Joshua Jackai, founder of Demand Designer, in this guide.

2. Build the content inventory: what happens to every existing page
A content inventory lists every existing page with its traffic and conversion value, then assigns each one a decision: keep, merge, improve or retire. It also names the new pages the redesign must add and who writes them. Without it, pages get lost or rebuilt twice.
Owner: marketing lead and whoever writes copy
Time: 3 to 5 days for a site under 100 pages
Output: one sheet with a URL, a traffic figure, a decision and a copy owner per row
Decisions: keep, merge, improve, retire
Export the page list from your CMS and the page-level traffic from Google Analytics and Search Console. Sort by traffic and by leads. A page with little traffic but a high conversion rate is a keep, not a retire, and a page with many visits and no leads is a rewrite, not a delete.
Every retire or merge decision creates a redirect, so this sheet feeds item 7. Google's guidance on moving a site with URL changes says to map the current URLs to their corresponding new ones, redirect the old URLs to the new ones and monitor traffic on both (Google Search Central, industry source). Writing that mapping while you decide, not during launch week, is the cheap way to do it.
The copy column is the one to watch. Who writes each new page, and by what date, is the most common cause of a slipped timeline I see. I ask for the copy, or an agreed outline of it, before the first design day.
"I ask for the copy, or an agreed outline of it, before the first design day," said Joshua Jackai, founder of Demand Designer.

3. Collect the brand assets: what the designer needs on day one
Brand assets are the files and decisions the visual design starts from: logos in vector format, brand colours and fonts, photography and any existing guidelines. If an item is missing, say so up front, so defining it is scoped as part of the project instead of discovered halfway through.
Owner: marketing lead or brand owner
Time: 2 to 3 days
Output: a shared folder with vector logos, a colour and font list, photo originals and a guidelines note
Honest note: which parts of the old guidelines are still true
Send logos as SVG or vector PDF, not a screenshot lifted from the old site. A screenshot caps the quality of everything built from it.
Photography needs a decision: reuse what exists, commission new work, or licence stock. Fogg and colleagues asked users what shaped their judgments of a site's credibility, so choose images you can defend as yours; the look of the site was "mentioned most frequently, being present in 46.1% of the comments" (DUX 2003).
Speed of judgment matters as well. Lindgaard, Fernandes, Dudek and Brown found that people form visual-appeal judgments within 50 milliseconds (Behaviour and Information Technology, 2006). That is a laboratory result from rated screenshots, so it describes first impressions and not conversion. I still treat it as a reason to have the logo, colours and first-screen proof ready before design starts.
If no guidelines exist, that is fine. Defining the visual direction becomes the first design step, and it is better scoped on day one than found in week four.
The look of the site was "mentioned most frequently, being present in 46.1% of the comments," Fogg and colleagues reported (DUX 2003).
4. Gather access to every account: the item nobody wants to chase
Access means working logins or invitations for every system the site touches: the domain registrar, hosting and CMS, Google Analytics, Search Console, and any CRM or email platform the forms connect to. Collect them before kickoff, because a build can stall for days on one forgotten password.
Owner: whoever holds each account, often three or four different people
Time: 1 to 2 days
Output: a list of systems with the named owner and a confirmed working login for each
Check: send the invitation, then confirm the invited person can actually log in
Registrar access is the one most often lost. The person who bought the domain left the company years ago, or it sits on a personal card. In my projects, recovering a domain has taken longer than building a page, so test this one first.
Analytics and Search Console access give the performance baseline for item 1 and the speed baseline for the new site. Web Vitals publishes thresholds for a good experience: Largest Contentful Paint at or under 2.5 seconds, Interaction to Next Paint at or under 200 milliseconds and Cumulative Layout Shift at or under 0.1 (web.dev, industry source). Record where the old site stands against them, so you can show what changed.
The rule I follow: I send invitations, not shared passwords. Invitations name a person, can be revoked, and leave a record. They also show you quickly which accounts nobody can reach. Ask each owner to confirm access in writing, and keep the list somewhere the whole project team can see it, so a missing login is visible on day one and not discovered on build day.
"I send invitations, not shared passwords," said Joshua Jackai, founder of Demand Designer.

5. Name one decision-maker and a reply time: the cheapest way to save weeks
A redesign needs one named person who can approve work without a committee, and an agreed feedback turnaround measured in days. In one of my projects, review rounds of seven days instead of two stretched a four week plan to seven weeks. The reply time is a decision you can make before kickoff.
Owner: executive sponsor or the project owner
Time: 1 day, mostly spent getting agreement
Output: a named approver, a named backup and a written reply time, usually two business days
Rule: comments from others go to the approver, who consolidates them
When several people approve, opinions multiply while the brief stays the same, and each round tends to reopen what the last one settled. I route all comments through one approver for that reason.
Nidumolu studied 64 software projects and found that coordination run by authorized entities, such as a project manager or steering committee, reduced project uncertainty and residual performance risk. Day-to-day communication between users and the technical team had a direct positive effect on performance (Information Systems Research, 1995). A website is not a software project of that size, so treat the paper as direction. The practical match is clear: one authorized approver and short loops.
Ask for a backup approver too. The most common stall I see is the approver going on leave in the middle of a review round. Clique Studios reports projects running 30 to 50 percent over estimate when content is late or approvals are slow (Clique Studios, read 23 October 2026). That is an agency figure, not a study, and it points the same way as the paper below. Write the approver, the backup and the reply time into the kickoff document and send it to everyone who will give feedback.
"In one of my projects, review rounds of seven days instead of two stretched a four week plan to seven weeks," wrote Joshua Jackai, founder of Demand Designer, in this guide.
6. Set the launch date with buffer: planning for the schedule you will actually have
Fix a launch date and list anything it must not collide with: a campaign, a conference, a quarter-end, a holiday. Then add written buffer to that date, because people underestimate how long their own tasks take, and a redesign has dozens of them.
Owner: project owner
Time: 1 day
Output: a launch date, a list of collisions and a written buffer of at least one review round
Pair with: the timeline guide, which compares published redesign durations
Buehler, Griffin and Ross found that people underestimate their own task completion times even when they know past similar tasks took longer than planned (Journal of Personality and Social Psychology, 1994). Kahneman and Lovallo trace the cause to an inside view of the problem.
"Overly optimistic forecasts result from the adoption of an inside view of the problem, which anchors predictions on plans and scenarios," wrote Daniel Kahneman and Dan Lovallo (Management Science, 1993).
The remedy is to look at how long similar projects took, not at your plan. Webstacks puts a standard B2B redesign at three to four months (Webstacks, updated 21 October 2026), and Design Studio UI/UX puts the fastest realistic path at six to ten weeks (Design Studio UI/UX, 10 August 2026). My own range on fixed-price builds is two to eight weeks, and the Snappy Kraken build, from Figma design to HubSpot, took 21 days. Those are first-party figures, and they depended on copy arriving on time and one person approving.
I write the buffer into the plan as a number of days, and the agreed reply time with it. The papers document the planning bias; they do not test that fix. Also check whether the platform is changing. A migration pulls the migration guides into scope and adds time.

7. Plan redirects and tracking: what protects traffic and leads on launch day
Before design starts, agree two things: a redirect map from every old URL to its new home, and a tracking plan that names each conversion event and where it must show up. Test a form submission end to end before launch, with its source attached.
Owner: designer for the redirect map, analytics owner for tracking
Time: 2 to 3 days
Output: a redirect sheet, a list of conversion events and a test plan
Test: submit the form and confirm the lead appears in the CRM with its source
The redirect map starts from the content inventory in item 2. Every retired, merged or moved page gets an old URL, a new URL and a permanent redirect. Pages that carry search traffic or links are the ones to check first. Google's site move guidance covers the redirecting and the traffic monitoring that follows (Google Search Central, industry source).
Tracking needs the same care. List each event that counts as a conversion: form submit, booked call, download. Decide which tool records it and where the lead lands. A form that records only an email address gives sales little to work with. Add a speed check to the same plan: Nah found that web users tolerated about 2 seconds of waiting without feedback for a simple task (Behaviour and Information Technology, 2004), so I test the new templates on a phone before launch, not after.
I do not call a build finished until a test form submission shows up in the CRM with its source attached. Judge a redesign at day 90 against the baseline, not on launch day against a feeling. Search results settle slowly, so a week-one dip tells you little.
"Judge a redesign at day 90 against the baseline, not on launch day against a feeling," wrote Joshua Jackai, founder of Demand Designer, in this guide.
What this checklist does not cover: preparation only
This checklist covers preparation only. It does not cover design review, build quality assurance or launch-day checks, which the process guide and the timeline guide handle. After launch I compare the redirect map with Search Console and the Web Vitals results with the baseline recorded in item 1.
Which checklist item matters most for your situation: a comparison matrix
The right order depends on what is most likely to go wrong in your project. The matrix shows each item against the risk it removes, so you can start with the one your team is least ready for in your own project.
Checklist item | Risk it removes | Owner | Time to gather | Start first if |
|---|---|---|---|---|
1. Goal and baseline | Judged on taste | Marketing lead | 1 to 2 days | Stakeholders disagree on success |
2. Content inventory | Lost or doubled pages | Marketing lead and writer | 3 to 5 days | The site has more than 50 pages |
3. Brand assets | Design starts from a screenshot | Brand owner | 2 to 3 days | Files live with a former agency |
4. Access | Build waits on a password | Account owners | 1 to 2 days | The domain owner has left |
5. Decision-maker | Review rounds stretch | Executive sponsor | 1 day | Three or more people approve |
6. Launch date | A date without a plan | Project owner | 1 day | A campaign or event is fixed |
7. Redirects and tracking | Traffic and leads vanish | Designer and analytics owner | 2 to 3 days | The site earns organic leads |
How ready are you: a readiness score you can run in two minutes
Score one point for each of the seven items you can show as a document, a login or a named person. Six or seven means you can start a conversation this week. Four or five means one or two weeks of gathering. Three or fewer means gather first, because starting now will cost more than waiting.
Published checklists run from 6 steps (Delta, 2024) to 23 items (Buarich), so seven is a deliberately short list. The scoring rule is mine and comes from client work, not a study, so read the bands as a rule of thumb. The count itself is simple:
6 to 7 items ready: book the scoping call and start.
4 to 5 items ready: spend one to two weeks gathering the missing items, starting with the ones that have the longest time to gather.
0 to 3 items ready: do items 1, 4 and 5 first. They unblock everything else.
In my projects, teams that arrive with the kit ready save about a week of timeline and a full round of revisions. That is my own observation across client work and not a controlled measurement.
Frequently asked questions about the website redesign checklist
What if I do not have brand guidelines?
If you do not have brand guidelines, defining the visual direction becomes the first design step, and I scope and price it as part of the project. Say so up front. Send whatever exists: logo files, colours in use and photography.
Do I need all the copy written before we start?
No, but you need to know who is writing it and when it lands. In my projects, late copy is the most common cause of a stalled redesign. An agreed outline for each page is enough to begin design while the final text is still being written.
Who should be the decision-maker?
One person with authority and availability, plus a named backup. Everyone else gives feedback to that person, who consolidates it into one set of comments. Coordination run by authorized entities reduced project uncertainty in Nidumolu's study of 64 software projects, so settle this before kickoff.
What should be on a website redesign checklist?
Seven things: a goal with a baseline metric, a content inventory, brand assets, access to every account, one named decision-maker and reply time, a launch date with buffer, and a redirect and tracking plan. Each takes one to five days to gather, and the longest is the content inventory.
What should be on a website redesign checklist for SEO?
Page-level traffic, a keep, merge, improve or retire decision for each URL, and a redirect map from every old address to its new one. Google's site move guidance asks for the old-to-new mapping and traffic monitoring. Record the organic baseline too, so day 90 has a comparison.
Who sets this up with you: how I run the preparation week
I run the preparation week as the first stage of a fixed-price B2B redesign: we go through the seven items together, I flag what is missing, and the plan names an owner and a date for each gap. For Snappy Kraken, the build from Figma design to HubSpot took 21 days (first-party client result).
If you want to see where your current site leaks before you prepare anything, the free Pipeline Leak Scorecard takes about fifteen minutes. Once the list is assembled, the remaining questions are cost and sequence. The redesign cost guide, the process guide and the timeline guide cover both, and the B2B website redesign guide shows where each fits. You do not need every box ticked to book a free scoping call. Bring what you have and I will tell you what is still missing.
Sources
Peer-reviewed research
Kohavi, R., Longbotham, R., Sommerfield, D., Henne, R. (2009). Controlled experiments on the web: survey and practical guide. Data Mining and Knowledge Discovery. doi.org/10.1007/s10618-008-0114-1
Nidumolu, S. (1995). The effect of coordination and uncertainty on software project performance: residual performance risk as an intervening variable. Information Systems Research. doi.org/10.1287/isre.6.3.191
Buehler, R., Griffin, D., Ross, M. (1994). Exploring the "planning fallacy": why people underestimate their task completion times. Journal of Personality and Social Psychology. doi.org/10.1037/0022-3514.67.3.366
Kahneman, D., Lovallo, D. (1993). Timid choices and bold forecasts: a cognitive perspective on risk taking. Management Science. doi.org/10.1287/mnsc.39.1.17
Nah, F. (2004). A study on tolerable waiting time: how long are Web users willing to wait? Behaviour and Information Technology. doi.org/10.1080/01449290410001669914
Lindgaard, G., Fernandes, G., Dudek, C., Brown, J. (2006). Attention web designers: you have 50 milliseconds to make a good first impression! Behaviour and Information Technology, 25(2). doi.org/10.1080/01449290500330448
Fogg, B. J., Soohoo, C., Danielson, D., Marable, L. (2003). How do users evaluate the credibility of Web sites? Proceedings of the 2003 Conference on Designing for User Experiences. doi.org/10.1145/997078.997097
Industry, vendor and first-party
Google Search Central, site moves with URL changes, read 3 November 2026 (industry source, not peer reviewed). developers.google.com
web.dev, Web Vitals thresholds, read 3 November 2026 (industry source, not peer reviewed). web.dev
Webstacks, How long should a website redesign take, updated 21 October 2026 (vendor source, not peer reviewed). webstacks.com
Clique Studios, Website redesign timeline and process, read 23 October 2026 (agency source, not peer reviewed). cliquestudios.com
Design Studio UI/UX, How long does a website redesign take, published 10 August 2026 (agency source, not peer reviewed). designstudiouiux.com
Delta, website redesign checklist: 6 steps (2024) and Buarich, website redesign checklist: 23 things, read 3 November 2026 (agency sources used for the 6 to 23 item range, not peer reviewed). godelta.com, buarich.com
Demand Designer timelines, the Snappy Kraken build and the review-round example are from my own client work, October 2026 (first party, not peer reviewed). demanddesigner.com/pricing
Free workbook
Pipeline Leak Scorecard
Find the leaks between your traffic and your pipeline. Score your site in about fifteen minutes, with it open in front of you.
12 questions, 8 pages, 15 minutes. Free.







