The Year-End SEO Review Every Law Firm Should Run Before December

Most law firms have the same December conversation. The agency sends a year in review deck, the numbers look encouraging, someone asks a hard question about intake, the room agrees to revisit it in the new year, and the contract renews on autopilot. Nobody is lying. The problem is that the only version of the year on the table was written by the party being evaluated.

The fix is not a bigger report. It is a small, dated, repeatable review that your own team runs before the vendor conversation happens, using data your firm already owns. Done properly it takes one person four to six hours spread across two weeks, needs no agency involvement, and produces a one page summary that changes the tone of the December meeting from presentation to comparison.

Below is that review, written so a marketing coordinator or an office administrator can execute it without an SEO background. Run it in the first two weeks of November. That timing matters. It gives you a full ten months of the year in the data, leaves room to ask follow up questions, and lands before renewal paperwork starts moving.

What This Review Is Not

This is not a technical audit. You are not going to crawl the site, evaluate schema markup, or judge the quality of a backlink profile. Those require tools and judgment your firm probably does not have in house, and they are the substance of what a real law firm SEO audit covers. This review is narrower and more useful. It answers five questions with numbers rather than opinions.

• Are we more visible than we were a year ago, for the cases we actually want.
• Can Google see and index the pages we paid to have built.
• How many real inquiries did the website produce, and what did each one cost.
• Which pages that used to work have quietly stopped working.
• Did the vendor ship what the agreement says they would ship.

Nothing here requires you to have an opinion about SEO. It requires you to be able to read a spreadsheet and count.

Gather Five Logins Before You Start

The single most common reason this review stalls is access. Ask for it in October, in writing, and treat resistance as a finding in its own right.

• Google Search Console. Your firm should hold owner level access, not delegated user access granted by the agency. If the property is verified under an agency email, that is a problem worth fixing this year regardless of who you work with.
• Google Analytics 4. Administrator access on the property, plus confirmation of which events are marked as key events.
• Google Business Profile. Primary owner should be a firm email address. Managers can be added and removed.
• The website admin. An administrator account in WordPress or whatever the CMS is, so you can see the publishing history.
• Call tracking and intake. Whatever records inbound calls and form submissions, plus the case management system that records what happened to them.

If any of these five are held exclusively by a vendor, write that down. Ownership of marketing data is not a technical detail. It is the difference between switching providers in three weeks and rebuilding two years of history from nothing.

Step One. Rankings You Actually Care About

Ignore any ranking report that leads with a count of keywords in the top ten. That number is trivially inflated by adding low volume terms nobody searches. Go to Search Console instead, open the Performance report, and set the date comparison to the last twelve months against the previous twelve months.

Now build a short list of the queries that map to cases you want more of. For most firms that is somewhere between fifteen and forty terms, and they are usually a practice area plus a city, a practice area plus a modifier like cost or how long, and the firm name itself. Filter the report to each one and record four columns. Clicks this year, clicks last year, average position this year, average position last year.

Read the result honestly. A move from position 14 to position 9 on a high intent term is worth more than a move from 4 to 2 on an informational one, because the click curve is steep at the top and nearly flat below the fold. Branded queries growing while everything else is flat usually means your other marketing is working and your SEO is not. And if impressions climbed while clicks did not, you gained visibility on terms that do not convert, which is a content targeting problem rather than a ranking problem.

Search Console holds sixteen months of data and the interface exports a maximum of one thousand rows at a time. For a firm with a large site, pull the data through Looker Studio or the bulk export instead of clicking through the interface. If that is beyond your comfort level, the fifteen to forty term manual version is entirely sufficient for this purpose.

Step Two. Indexation and Crawl Health

This step catches the most expensive failure in law firm SEO, which is paying for pages that Google never puts in its index. Open the Pages report in Search Console. You will see two totals, indexed and not indexed, with reasons listed underneath.

Compare the indexed count against the number of pages your site actually has. Your sitemap gives you that number. A gap of a few percent is normal. A gap of thirty percent means a meaningful share of what you commissioned is invisible. Then read the reasons. Crawled currently not indexed usually signals thin or duplicative pages, which is common when a vendor has mass produced dozens of near identical city pages. Duplicate without user selected canonical points at template problems. Excluded by noindex tag on a page you want ranked is a straight configuration error, and it happens more often than anyone admits, usually after a redesign when a staging setting ships to production.

Spot check ten pages you consider important by pasting each URL into the Search Console inspection tool. Record indexed or not indexed for each. Ten minutes of this reliably finds something.

Step Three. Conversions and What They Cost

Traffic is an input. Signed cases are the output. Most firms never connect the two, which is why the December conversation drifts into abstractions.

In GA4, pull key events for the year with organic search as the channel. You want form submissions, phone clicks, and chat starts separated, not lumped into a single conversions figure. Then get the real numbers from intake for the same period. How many inbound inquiries, how many consultations booked, how many signed. Note that GA4 defaults to a fourteen month data retention window for the exploration reports, so if nobody set it to the longer option you may be missing history. Fix that setting today, it takes thirty seconds and cannot be applied retroactively.

The gap between analytics and intake is always instructive. Analytics counts a phone number click on a mobile device. It does not know whether the call connected, whether the person was in your state, or whether the matter was one you take. If GA4 reports 240 conversions and intake logged 90 usable inquiries, that ratio is your reality, and everyone should be working from the second number.

Then divide. Annual SEO spend divided by signed cases from organic gives you a cost per case. Compare it against what you pay for the same case from paid search or referrals. This one figure reframes the entire renewal discussion, and it is the number most vendor reports carefully avoid producing. If you want to understand the reporting habits that obscure it, we covered them in how to read the SEO report your agency sends.

Step Four. Content Decay

Content does not fail loudly. A practice area page that brought 400 visits a month two years ago drifts to 260, then 150, and no report ever flags it because the site total looks stable.

Find the decay by going back to the Search Console Performance report, switching to the Pages tab, and running the same year over year comparison. Sort by the difference column, ascending, so the biggest losers surface first. Take the top twenty and put them in a spreadsheet with three columns. What it was, what it is now, and a decision.

There are only four honest decisions for a decayed page.

• Refresh. The page still targets something you want. Update the law, the numbers, the examples, and the internal links, then republish. This is the right call for most practice area pages.
• Consolidate. Three overlapping pages are competing with each other. Merge into the strongest one and redirect the others.
• Redirect. The page is obsolete but has links or history. Point it at the closest relevant page with a 301.
• Leave it. It declined because demand declined, or it was never a page you cared about. Not everything needs attention.

What you must not do is delete in bulk because a tool called the pages low quality. That advice has cost firms real rankings. Decisions are made per URL, with a reason recorded for each. Doing this properly at scale is the discipline described in content inventory and topic mapping for law firms, and the annual review is a good moment to check whether that inventory exists at all.

Step Five. Local Presence and Reviews

For firms competing on geography, the map pack often produces more calls than the organic listings below it. Open Google Business Profile and pull the performance data for the year. Record calls, direction requests, website clicks, and the split between searches that found you by name and searches that found you by service.

Then check the unglamorous things. Are the hours right, including the holiday closures you are about to have. Does the address and suite number match exactly what appears in your website footer and on your state bar listing. How many reviews did you add this year, and did they arrive steadily or in two suspicious clusters. A firm adding two or three reviews a month consistently outperforms one that added thirty in March and nothing since, and the review text itself increasingly matters because it is source material for the AI generated summaries now sitting above the traditional results. The ABA Journal’s coverage of how firms are adapting to generative engine optimization in AI search is a reasonable primer if your partners want outside reading on why this shift is real.

Step Six. Vendor Performance on Paper

This step needs no analytics at all. Open the agreement or the last proposal, list every deliverable it names with the promised frequency, and then count what actually arrived.

• Pages or posts published. Count them in the CMS by publish date. Compare against the promised cadence.
• Technical fixes. Ask for the list of what was changed and when, then verify two or three of them yourself.
• Links or citations. Ask for the specific URLs acquired this year, not a count. If the answer is a count, that is the finding.
• Reporting. Did it arrive monthly as agreed, and did it ever contain a recommendation the firm was asked to act on.
• Meetings. How many were held, how many were rescheduled by the vendor.

A serious partner will have no trouble producing all of this within a week and may well volunteer the misses before you ask. A vendor who responds to a deliverable audit with a new dashboard has told you something important.

Step Seven. The Ten Minute Risk Scan

Finally, check the things that only become emergencies when nobody looks.

• Domain registration expiry date and who the registrant is. Renew anything expiring within twelve months.
• SSL certificate expiry.
• Whether attorney advertising disclaimers required in your jurisdiction still appear on the pages that need them, especially on newer pages a vendor added without review.
• Whether any page makes a results claim your state bar would treat as a comparison or a guarantee.
• Whether staff who left the firm still appear on the site, and whether their bio pages return errors or redirect somewhere sensible.
• Who has admin access to the website, and whether any of those accounts belong to former employees or former vendors.

Turning It Into One Page

Do not walk into December with a forty tab spreadsheet. Reduce everything above to a single page with three columns. What we measured, this year against last year, and the question it raises. Six to ten rows is plenty.

Then send that page to the vendor a week before the meeting. This is the part firms hesitate over and it is the part that works. It converts the meeting from a performance into a working session, gives a good partner the chance to arrive with explanations and a plan, and gives a weak one nowhere to hide. Either outcome is worth knowing in December rather than the following August.

One caution on interpretation. A single year of data will not tell you whether a vendor is good, only whether the numbers moved. Search results shift for reasons no agency controls, including algorithm updates, a competitor’s acquisition, or the fact that your practice area’s demand genuinely fell. The right posture is to ask what changed and whether the explanation is specific and checkable. Vague answers about updates and volatility, repeated across two consecutive years, are their own answer.

Make It Annual

The value of this review compounds. The first year it produces a snapshot. The third year it produces a trend line that no vendor deck can talk you out of, because you built it yourself from the same sources in the same order every November.

Put it on the calendar now. Access requests go out October 1. Data pull runs the first week of November. One page summary is written the second week. Vendor receives it the third week. The meeting happens in December, and it starts from your numbers instead of theirs.

Rubiks works with law firms that arrive holding exactly this kind of document, and we would rather be evaluated against a firm’s own numbers than our own slides. If you run the review and find gaps you cannot explain, whether that is a third of your pages missing from the index, a cost per case nobody has ever calculated, or a decay list nobody has looked at in two years, Talk to Rubiks and we will read your findings with you and tell you plainly which items are worth budget in the coming year and which ones are not.