The Website Scorecard
This scorecard tracks whether the website is turning visitors into leads and whether the site itself is healthy enough to keep doing it. It covers key-page traffic, form and call conversion by page, page speed, tracking health, broken links, content freshness, and how fast someone follows up after a form submit. The web lead runs it, with the developer, content lead, and whoever answers leads each owning a row.
Why this scorecard exists
Why the website needs its own scorecard
Most website reporting is a traffic chart. Sessions went up, sessions went down, and nobody can say what that did to the pipeline. A generic marketing scorecard makes it worse by rolling the site into one conversion rate for the whole domain, which hides the fact that the service page converts and the blog does not. The site is a set of pages with different jobs, and a scorecard that cannot see the difference cannot tell you what to fix.
The website is also the only channel where the measurement layer can break silently. A tag manager update, a form plugin upgrade, or a theme change can stop conversion events from firing, and every ad platform keeps optimizing against the wrong signal for weeks before anyone notices. Speed drifts as pages get heavier. Links rot as pages get renamed. None of that shows up in a traffic chart, so it needs rows of its own, with someone checking them on a schedule instead of finding out from a client.
When this scorecard is live, the weekly conversation changes from "how was traffic" to "which page slipped, why, and who is fixing it." A drop in form conversion on one page becomes a ticket with a name on it. A tracking failure gets caught the day it happens instead of the month after. Lead response time becomes a number the sales side owns, not a complaint marketing raises. The site stops being a thing that gets redesigned every three years and becomes a system someone maintains every week.
The scorecard
10 numbers. Every one has an owner.
| Metric | What it means | How to set the target | Owner | Cadence | Red flag |
|---|---|---|---|---|---|
| Visits to key pages Total sessions hide the pages that matter; a drop on the service page is a problem even when the blog is up. | Sessions that land on or reach the handful of pages whose job is to produce a lead, such as service pages, location pages, pricing, and contact, tracked page by page. | Set the floor per page at its trailing 8-week median, then raise it only when a page has beaten the median for four straight weeks. Tie any new target to a specific change, such as a new campaign or a content update. | Web lead | Weekly | Any key page down two consecutive weeks against its median starts a source review to find which channel dropped. |
| Form conversion rate by page One page with a broken layout or a weak offer can drag the site average down while every other page is fine. | Completed form submissions divided by sessions, calculated separately for each key page rather than for the whole site. | Use each page's own trailing 8-week median as the baseline and tighten it by a set step each quarter after a test wins. Compare pages to their own history, not to each other, since a contact page and a service page will never convert the same. | Web lead | Weekly | A key page falling below its median for two consecutive weeks triggers a test submission and a review of the last deploy. |
| Call conversion rate by page For many businesses the phone outproduces the form, and a page that loses calls but keeps form fills still loses leads. | Tracked phone calls, from click-to-call taps and tracked numbers, divided by sessions on each key page. | Baseline each page against its trailing 8-week call rate, with mobile and desktop split, since calls come mostly from phones. Set the target as the median and revisit it only when the tracking setup changes. | Web lead | Weekly | A drop in call rate on any page that traffic does not explain triggers a check that the tracking numbers still swap correctly. |
| LCP on key pages Slow pages lose visitors before they see the offer, and speed drifts every time someone adds an image or a script. | Largest Contentful Paint, the time it takes the biggest visible element to load, measured on mobile for each key page. | Record each key page's LCP after a clean build and hold it there, allowing no more than a small agreed drift. Set the mobile number as the target since that is where most first visits happen. | Developer | Weekly | Any key page slower than its baseline by the agreed drift for two weeks opens a performance ticket before any new content ships. |
| Tracking health When events stop firing, every paid campaign optimizes against nothing and the scorecard lies for weeks. | A weekly count of conversion events, such as form submits and call clicks, that fired and reached analytics and the ad platforms as expected. | The target is a pass on every event in the checklist, verified by a scripted test submission each week. Compare the event count in analytics to the form tool's own submission count and expect them to match within a small, agreed margin. | Developer | Weekly | Any event missing from the test run, or a gap between form-tool submissions and analytics events beyond the agreed margin, is fixed the same day and logged. |
| Broken links and 404s Broken links waste clicks, hurt search rankings, and usually mean someone renamed a page without a redirect. | The count of internal links that lead to a missing page, plus 404 hits recorded in server or analytics logs, from a weekly crawl. | The target is zero broken internal links on key pages and a 404 count at or below the trailing 8-week median across the site. Every rename or deletion should ship with a redirect, so the number should only go down. | Developer | Weekly | Any broken link on a key page, or a 404 count above median two weeks running, gets a redirect ticket before the next review. |
| Content freshness Stale prices, old team photos, and outdated offers cost conversions and signal neglect to search engines and visitors. | The number of key pages whose content, offers, and proof were reviewed and updated within the agreed window. | Set a review window per page type based on how fast its content changes, then track the share of key pages reviewed on time. Start with the window your team can hit every month and tighten it once it holds. | Content lead | Monthly | Any key page past its review window by more than a month is scheduled for an update in that week's review. |
| Lead response time A fast site that feeds a slow inbox still loses the lead, and this is the row where marketing and sales meet. | The median time between a form submission and the first human reply by phone, text, or email, measured from the form timestamp. | Pull the last eight weeks of form timestamps and first-reply timestamps to find your current median, then set the target below it in steps. Track business hours and after hours separately so the number reflects what the team can control. | Sales lead | Daily | Any lead without a first reply inside the agreed window is escalated to the owner the same day. |
| Form delivery check Forms fail quietly when an email address changes or a CRM integration expires, and the visitor never knows. | A weekly test submission on each key form, confirming the lead reached the inbox and the CRM with the right source fields attached. | The target is a pass on every form every week, with the test lead landing in the CRM under the correct source and page fields. Anything less than a full pass counts as a fail for the week. | Web lead | Weekly | A test lead that does not reach the CRM within the hour halts any other site work until the integration is fixed. |
| Key pages indexed A page that fell out of the index gets no organic traffic no matter how well it converts. | The count of key pages that search engines report as indexed and free of errors, checked against the list of pages that should be. | The target is every key page indexed with no errors, so the number is a simple match against your page list. Any gap is a defect, not a trend. | Web lead | Monthly | Any key page missing from the index or flagged with an error opens a ticket to check its robots rules, canonical tag, and redirects. |
Targets are set from your own history, never from someone else's benchmarks. The builder does the arithmetic once you plug in your numbers.
The weekly review
Twenty minutes, same order, every week.
// 01
Start with conversion by page
Put the form and call conversion rows up first, page by page, against each page's median. This is the number that ties the site to the pipeline. The web lead calls out any page that is red and states whether traffic, tracking, or the page itself is the likely cause.
// 02
Check the health rows before trusting the rest
Look at tracking health, form delivery, and broken links next. If any of those failed this week, the conversion numbers above are suspect and the fix goes to the top of the list. The developer speaks here and confirms what was fixed and what is still open.
// 03
Look at speed, freshness, and traffic
Scan LCP, content freshness, and key-page visits for trends rather than single-week dips. The content lead names any page past its review window. The web lead flags any traffic drop that a channel change explains, so it does not get treated as a site problem.
// 04
Close on lead response and decide
The sales lead reports median response time and any lead that missed the window. The group then leaves with a short list: which page gets a fix, which ticket the developer owns, and which content update ships this week. Every red row leaves the room with a name and a date.
Where scorecards die
Mistakes we see constantly.
Reporting one site-wide conversion rate
A single number for the whole domain averages the service page with the blog and the careers page, so nothing moves it and nothing explains it. Track conversion on the pages that are built to convert, and track them separately.
Assuming the tracking still works
Events fail quietly after plugin updates, theme changes, and consent banner tweaks, and the scorecard keeps reporting numbers that are no longer true. A weekly test submission takes five minutes and catches most of it.
Treating speed as a launch-day task
The site was fast the day it shipped and then someone added a hero video, three tracking scripts, and a chat widget. LCP belongs on the scorecard so drift gets caught while it is still small.
Stopping the scorecard at the submit button
The site did its job, the form fired, and the lead sat in an inbox for two days. Lead response time belongs on the website scorecard because that is the point where the work gets wasted.
Chasing traffic with no page owner
Traffic goes up, conversion goes down, and nobody owns the page where it happened. Every key page needs an owner who answers for its row, or the review becomes a report nobody acts on.
Other scorecards
Same discipline, different numbers.
By platform
By business type
Want it set up live in your dashboards?
We install scorecards on real data, wire the owners and alerts, and coach the weekly review. Scorecards and accountability is the service.
Questions we get
Straight answers.
Which pages count as key pages?
The pages whose job is to produce a lead or a sale: service pages, location pages, pricing, contact, and any landing page paid traffic hits. Keep the list short enough that every page can have a named owner. The home page usually belongs on it; most blog posts do not.
Where does the data come from?
Analytics for visits and conversion events, your form tool and CRM for submissions and response time, a call tracking tool for calls, a speed tool for LCP, a crawler for broken links, and Search Console for indexing. The scorecard pulls one number from each. The point is not the tools but that each row has one agreed source.
How do we set targets without an industry benchmark?
Use your own history. Pull the trailing eight weeks for each row, take the median, and make that the floor. Tighten it in small steps only after a change proves it can hold, so the target reflects what your site actually does rather than a number from someone else's site.
Who should run the weekly review?
The web lead, or whoever is closest to the site day to day. The developer, content lead, and sales lead each own rows and speak to them. The owner attends when something is red two weeks running or when a decision needs budget.
How does this connect to the paid media and SEO scorecards?
Paid media sends the traffic and this scorecard measures what the landing pages do with it, so the two share the tracking health row. SEO owns rankings and organic traffic, and this scorecard picks up at the page. If conversion by page is red, the paid and SEO teams need to know before they change anything upstream.
A scorecard is only as good as the system feeding it.
Start with a diagnostic call. We will tell you which of these numbers you can trust today and what it takes to trust the rest.