SEO Proposal for Client PDF

I start with a real audit, not a template. Before I open a proposal document I pull data from Google Search Console and a tool like [Semrush](/semrush/) or [Ahrefs](/ahrefs/). I look for three things: traffic trend over six months, the top 20 keywords by clicks, and any technical errors flagged by a [Screaming Frog](/screaming-frog/) crawl. For a recent ecommerce client who sold climbing gear, I found they had 2,000 indexed pages but 80% of traffic went to six product pages. Their bounce rate was 67% on category pages. That told me the strategy had to fix internal linking and category copy before anything else. I opened the proposal with that snapshot: "Your site attracts 14,000 visitors a month but 11,000 leave from category pages. Here's why."

Next I write the client overview and objective statement. I keep it short: one paragraph on their business, one on what they told me they want (usually more traffic or sales), and one on what I think is actually achievable in six months. I use numbers from the audit to anchor expectations. For the climbing gear client I said: "Your organic traffic has grown 8% year on year, but your main competitor has grown 34%. I believe we can close that gap to within 10% by month six." That is a concrete claim I can be held to. The margin for error comes from the uncertainty of algorithm shifts, which I flag in the next section.

The strategy section is the meat. I split it into four workstreams: keyword and content, on-page optimisation, technical SEO, and link building. For each I list two or three specific tasks and tie each to a metric. For example: "Optimise 15 category pages with unique copy and internal links — target: reduce bounce rate to under 50%." I also include a timeline, typically a Gantt-style table by month. Month one is audit and quick wins; months two to four are core optimisation; months five and six are link building and refinement. Pricing sits right after the timeline. I charge a flat monthly retainer based on the number of workstreams active. I never bill by the hour because that creates a perverse incentive. For the climbing client the retainer was £2,000 per month for six months, with a clause that we reassess after month three.

I close the proposal with a one-page summary of deliverables, a contact section, and a call to action: a 30-minute call to walk through the proposal. I send it as a PDF, never a Google Doc, because formatting breaks. The file name is the client's name and the date, not "SEO Proposal Template."

The most common mistake is writing a proposal that could fit any business. If you replace the client name and nothing else looks wrong, you have not done the work. I made that error twice before I started forcing myself to reference client-specific data in every section. A proposal that says "Your site has technical issues" without listing any is useless. Name the issue: "Your homepage loads in 4.2 seconds on mobile, which we measured with a [Website Speed Test](/website-speed-test/). The target is under 2.5 seconds."

Second watch-out: listing tasks without connecting them to outcomes. Clients do not buy keyword research; they buy the traffic that research unlocks. For every task I write a short line explaining the business impact. For example: "We will conduct a competitor gap analysis using [Similarweb](/similarweb/) to identify keywords where you have zero visibility but your competitor gets 500 visits a month. Capturing even 10% of that traffic adds £X in potential revenue at your conversion rate." That turns a task into a reason to pay.

Third: overpromising. Never write "We guarantee page-one rankings" or "We will get you 10,000 new visitors in three months." Google does not guarantee anything. Instead say: "Based on our audit, we expect to move your top 10 keywords from an average position of 18 to position 6 within six months, given consistent effort and no major algorithm changes." That is honest and gives the client a realistic benchmark. If they push for a guarantee, I explain that no reputable SEO practitioner offers one and cite Google Search Central's guidance on quality expectations. If they still want guarantees, I pass on the project.

Fourth: ignoring the local dimension when it matters. If the client depends on foot traffic, a proposal without [Brightlocal](/brightlocal/) data or Google Business Profile recommendations is incomplete. For a chain of three coffee shops, I once wrote a proposal that focused purely on blog content. I lost the deal to someone who showed them how to optimise GBP listings and local citations. Now I always check the client's local presence in the audit phase.

Fifth: too much jargon. I avoid terms like "canonicalisation" or "TF-IDF" in the proposal. Instead I say "we will make sure Google does not index multiple versions of the same page" and "we will create content that naturally matches the language your customers use." Every technical term I include gets a one-sentence explanation because the person signing the cheque may not be an SEO.

Edge case: when client has zero traffic data (new site or no analytics). In that case I use industry benchmarks from Similarweb and competitor data. I also recommend installing [Microsoft Clarity](/microsoft-clarity/) or Hotjar as a first step to start collecting user behaviour data. The proposal then focuses on setting up a baseline rather than promising results from nothing.

Early in my career, every proposal I wrote was a feature list. "We will do keyword research. We will optimise meta tags. We will build links." I never explained why those features mattered to the client's bottom line. I remember one loss vividly: a B2B SaaS company that wanted to grow demo requests. I sent a seven-page proposal packed with tasks but no connection to their goal. A competitor sent three pages that started with "You currently get 120 demo requests a month from organic. The top search result for 'project management software UK' gets 8,000 visits a month. Here's how we get your page into the top 5." They won the deal. That taught me that clients buy outcomes, not activities. Now every proposal starts with a problem statement pulled from their actual data.

I also used to leave out the audit. I thought I could write a good enough proposal from a quick glance at their site. But without a proper crawl using Screaming Frog or a backlink profile check via Ahrefs, I was guessing. I once proposed a huge content programme for a site that turned out to have 200 broken internal links and a 404 rate of 12%. The technical fix would have given them a 30% traffic lift faster than any content. The client rightfully asked why I did not catch that earlier. Now I never propose a strategy without first running a full site health check.

Another thing I changed my mind about is pricing presentation. I used to bundle everything into one retainer without a breakdown. That made it impossible for the client to see what they were paying for. Now I provide a table: Month 1 audit and quick fixes (£X), Months 2-3 content and optimisation (£X per month), Months 4-6 link building and monitoring (£X per month). If the client wants to drop a phase, they can see the trade-off. It builds trust.

I am still uncertain about how much technical detail to include. Some clients want to know exactly what happens under the hood; others just want the traffic. I have not found a universal rule. My current heuristic: if the marketing contact is the decision-maker, keep the technical detail to 20% of the proposal. If an in-house developer or CTO is involved, bump it to 40%. I flag this uncertainty in the kickoff call and offer to add an appendix if needed. It is not perfect, but it stops me from either overwhelming or underwhelming the reader.

Finally, I used to send the proposal as a Word document. The formatting always broke. Now it is a PDF with the client name in the file name. Small thing, but it signals care.