The SEO proposal landed in your inbox as a PDF, twenty pages or so, your logo on the cover and a price on the last page. You read the last page first. Then you skimmed the audit, saw a lot of red, and reached the end without finding the one thing you opened it to learn: what exactly happens in the first month if you say yes.
I write proposals and I read them. Killerspots, the agency behind KillerRank.ai, has been sending them since 1999, and owners forward me the ones they get from other companies and ask what they are looking at. So this is a reading guide, not a writing guide. What each page is really for, the page that is almost always missing, how to tell a template from a document written about your business, what the promise words mean once you translate them, and the three questions to send back before you get on the call.
What is an SEO proposal actually for?
Start by being honest about what the document is. Somebody built it to close. That is not a criticism, it is the genre, and the agency side of our own company writes them the same way. Our sister site for agencies has a post on the SEO proposal template from the writer’s chair. If you want to know what the person who sent yours was thinking, read it. This post is the other chair.
The writer is paid when you sign, so the document is built around the moment of signing: the gap, the urgency, the recommended package in the middle column. You are paid when the work happens, so you need it to describe month two, month six and the month something goes flat. Almost no proposal is written for those months, which is why the page you most need is usually the one that is not there.
What pages does every SEO proposal have, and what is each one really for?
The audit screenshots. Red scores, a crawl error count, a list of missing descriptions, a chart of your traffic next to a competitor’s. This page exists to make the problem feel real, and often it is. Two checks. First, are the findings true on your live site when you open it logged out? Audits run with a browser plugin or a tool account pick up things no customer ever sees, and I have watched a report flag eight problems on a page where only five existed in public. Second, does the plan on the later pages fix the findings on this one? A problem raised with no fix attached is decoration. Google’s own page on whether you need an SEO says an audit should give you “realistic estimates of improvement, and an estimate of the work involved”, and that is a fair bar for the free one in a proposal too. The SEO audit cost post covers what a paid audit should contain.
The strategy and keyword list. Check who picked the words. A list of the searches you want to win is a plan. A list of searches with big volume numbers next to them is a tool export, and the ones with the biggest numbers are usually the ones nobody small can win. Look for the searches a buyer types with a town or a job in them, and look for your own service pages named as the pages that will rank for them. If no page of yours is named, the list is not attached to your site yet.
The timeline. Month one, foundation. Months two and three, build. Months four to six, momentum. Every proposal has a version of this, and the words are not wrong. The check is whether any month contains a noun. “Technical fixes and content strategy” is not a deliverable. “Four posts, two of them for your two main services, the redirect map for the old addresses, Search Console connected under your account” is. A timeline with no nouns in it is a promise to be busy.
The package table. Three columns, the middle one marked recommended, prices across the bottom, built so you choose the middle. That is fine as long as the rows can be counted. The SEO packages post teaches the line test for every row: what is it, how many, and where will I see it. The cost post shows how to lay two tables from two companies side by side so the padded rows stand out. Do both before you look at the price again.
Which page is usually missing?
Here is the page I look for first, and the page I almost never find. It has two halves.
The first half is a short list of names. Who writes the posts that will carry your byline. Who decides what gets built next month, and from what data. Who you call when something is wrong. If the proposal says “our team” or “a dedicated account manager” and stops there, you do not yet know who does the work, and in my experience the answer is often a different company. The SEO consultant vs SEO company post is about exactly that question, and the hire an SEO expert, an agency or do it yourself post covers the routes the work can take, which is worth knowing when a proposal shows a plan with no people in it.
The second half is month one, written as a list of things that will exist by day thirty. Not “onboarding and strategy”. Posts, by topic. Pages, by name. Fixes, by page. The accounts that will be connected and under whose login. The day the first report lands. This list is what turns a proposal into something you can check, because on day thirty-one you can open the site and count.
Why is it missing? Usually one of three reasons: the plan gets built after you sign, the honest answer is thinner than the proposal implies, or it is a template and nobody filled that page in for you. None of those is a reason to walk away on its own. All three are a reason to ask, and the answer tells you more than the twenty pages did.
How can you tell a template proposal from one written for your business?
Templates are not a sin. The structure of a good proposal should be the same every time, and a company that rebuilds the format for each prospect is wasting hours you will pay for. The audit page and the strategy page are different. Those two have to be about you, and the swap test tells you in a minute whether they are.
Look for four kinds of specifics. Your pages, written as addresses, with a sentence about each that could only be true of that page. Your towns or states, in the order they matter to you, which only you could have told them. Your competitors as websites, not as “local competitors”, with the searches where they beat you. And a number from your own Search Console or Google Business Profile rather than a third party estimate, which means somebody asked you to read a number off the screen on the call.
One caveat from the other side of the desk. A good company will not ask for your analytics logins before you have agreed to anything, so a first proposal built from outside data is normal. The test is whether that data was read. A proposal that counts your pages and says how many of them earn no visits was built by somebody who looked. One that says your site “has significant technical and content opportunities” was built by somebody who did not.
What do the promise words translate to?
Every proposal has a vocabulary, chosen because it is hard to argue with. Nobody is against results. But you cannot check a result against an adjective, so before you compare two proposals, translate each promise word into the noun it would have to become to be checkable.
Results becomes the list of numbers that will be reported, where they come from, and the day each month they arrive. Growth becomes growth of a named number, from a dated starting point, which means the proposal has to state where you are today. Visibility becomes a list of the searches and the AI questions you will be tracked on, with the answer that currently names a competitor. Authority becomes links, and links become sites, named after they are live, pointed at a page you want to rank. The link building post covers what a real one looks like on an invoice.
Two words deserve a harder look. Guaranteed, in any form, is a sales device, and the guaranteed SEO services post covers what a company can honestly put in writing. “Pay for results” is a pricing model with its own fine print, covered in the pay for performance post. Neither changes the nouns. Translate those first, then check the price against the math in the is SEO worth it post.
What three questions should you send back before the call?
Send the questions by email before the call, not during it, because a proposal walkthrough is a performance and the answers you get live are the ones the presenter has rehearsed. Written answers tell you how the company works when nobody is watching.
The first question fills in the missing page. A good answer has names and nouns. A weak answer repeats the process diagram. The second question connects the audit to the plan, and it catches the padding: the finding that was there to scare you and will never be touched. The third question is the one that matters most and gets asked least. Month one almost always looks productive, because setup is visible. Month two is where you find out whether there is a plan or a retainer. A good answer describes what gets built in month two regardless of the numbers, because the numbers are not supposed to move yet. A bad answer talks about patience.
Then read the reply the same way you read a monthly SEO report. Work named, numbers dated, a person attached. If the proposal already answers all three and the price fits your math, sign it, whoever sent it. You do not need us for that. The companies that answer those questions well are the ones we would hire too.
What does our proposal look like?
This is the one place in this post where I am describing my own product, so I will be plain. We decided early that an owner should not need a sales call to learn what a plan contains or costs, so the plans page prints all of it. Starter is $997 a month and Growth is $1,997, with Authority at $2,997 and Dominate at $4,997 above them, and the setup fee is itemized and waived on every plan. Each column is counted: four, eight, twelve or sixteen posts a month, fifty to five hundred tracked keywords, sixty to two hundred fifty AI questions, and on Growth and up two, four or eight authority links and new pages a month. If a line cannot be counted, it is not on the page.
The missing page lives on the how it works page. We need three things from you: access to Search Console, Google Analytics and your Google Business Profile, the name of a real person at your company who will be the author on your posts, and about thirty minutes a month. The first posts go live inside two weeks of kickoff, after you have read them, under your author’s name and never ours. Every month after that is the same engine, and the written report and strategy call land whether the numbers moved or not. The done for you SEO post lists the month in full, and the results page shows dated positions for real clients, first check to latest check, so the proof is a window and not a best week.
The one thing the plans page cannot do is sit on the call with you. If you want that, ask us the same three questions. We will answer in writing, with names.
Before you reply
Read the proposal in this order, which is roughly the reverse of how it was written. The package table, with the line test on every row. The timeline, counting the nouns. The audit page, logged out, checking which findings the plan actually touches. Then look for the missing page, and when it is not there, send the three questions. The SEO contracts post picks up where this one stops, because the proposal promises and the contract binds, and the gap between them is where owners get hurt.
If you would like the audit page of a proposal without the proposal, the free audit reads your site the way Google and the AI tools do and lists what is missing, by page. The SEO company for small business page explains how we work the first month, and every plan runs on a seven day free trial, so the month one list is something you watch arrive rather than something you take on faith.
Questions We Get Asked
What should an SEO proposal include?
At minimum: findings about your site that are true on the live, logged out page; a list of the searches it will go after and who chose them; the work that ships in month one, named; who does that work; what the report shows and on what day; the price; and how you leave. A proposal missing the month one list and the people is a brochure with your logo on it.
How long should an SEO proposal be?
Long enough to name the work and short enough that you read the whole thing. The useful part of most proposals fits on two pages: the findings that the plan will fix, the month one list, the people, the price and the exit. The other fifteen pages are usually audit screenshots and a process diagram. Read those last, if at all.
Is an SEO proposal a contract?
No. A proposal is a sales document; the contract or terms of service is what binds either side. Anything the proposal promises that the contract does not repeat is not promised. Read the two together, and if the proposal names deliverables the contract leaves vague, ask for the deliverables to be written into the agreement before you sign.
Should I get SEO proposals from more than one company?
Yes, and send each company the same brief: your service pages, the towns or states you serve, the three competitors you lose to, and what a new customer is worth to you. Then compare the month one lists and the named people, not the prices. Three proposals built from the same brief are comparable. Three built from three different sales calls are not.
Is there an SEO proposal example I can compare mine against?
Use our plans page as the example. It prints every line of each plan, how many posts, pages, links, keywords and AI questions, with the price and the trial terms on the same page, and the how it works page lists what happens in the first two weeks. Hold any proposal next to that: if a line cannot be counted or a month cannot be named, ask why.