Most SEO proposals are difficult to compare because they describe activity rather than outcomes, and because the cheapest and most expensive can look identical on paper. Here is what to read for, and the specific promises that tell you the person writing it either cannot deliver or intends to take shortcuts.
The claims that should end the conversation
A guaranteed ranking
Nobody controls Google’s results. A guarantee of position one for a competitive term is either meaningless, because it will be for a phrase nobody searches, or it is a signal that the person intends to use tactics that work briefly.
The exception worth allowing: a guarantee tied to something they actually control, like delivery of specific work by a specific date.
Results in thirty days
Technical fixes can show up in weeks. Content and links compound over months. Anyone promising rankings in a month for a competitive term is describing paid advertising, not SEO, or is planning something that will cost you later.
A fixed number of links per month
Real link acquisition is lumpy, because it depends on other people deciding to link. A guaranteed monthly quantity means the links are being bought or placed, which is a violation and a liability attached to your domain. The links worth having cannot be promised by the dozen.
A secret method or a proprietary algorithm
There is no secret. The work is well documented and the difference between practitioners is judgment and execution, not access to hidden knowledge.
A fixed number of “SEO articles” per month
Content volume as a deliverable produces content nobody reads, written to a quota. Ask what each piece is supposed to achieve and for whom.
What a good proposal contains
Findings before recommendations. Anyone proposing work without having looked at your site is selling a package. A proposal should tell you something about your specific situation you did not already know.
A stated priority order, with reasons. Technical before content before links is the usual right order, because a site that cannot be crawled cannot rank regardless of what else is done. If the order is different, there should be a reason.
Specific deliverables. “Ongoing optimization” is not a deliverable. “Fix the crawl errors identified in section two, rewrite the four service pages, implement structured data” is.
The measurements, named in advance. Rankings for which terms, and more importantly, enquiries. If the proposal only measures rankings, it can succeed while your phone stays silent.
What they need from you. Every real program requires client input. A proposal that requires nothing from you is describing work that happens at arm’s length from your business.
What is excluded. Honest proposals say what is not covered.
Questions worth asking before signing
- What would make you tell me to stop spending money on this?
- If my website itself is the problem, does this proposal cover fixing it or does that become a separate project?
- Who owns the accounts, the content and the reporting if we part ways?
- What does month nine look like compared to month two?
- Have you ever told a client that SEO was the wrong investment for them?
The last one is the most useful. Anyone who has never advised against their own service has either been extraordinarily lucky or has not been looking.
On price
Cheap SEO is usually one of two things: automated work at a scale that cannot be personalised, or shortcuts that carry risk. Both look identical to good work for the first several months, which is exactly the problem.
Expensive is not automatically better either. What you are paying for is judgment and execution time, so the question worth asking is who is actually doing the work and how much of their time you are getting. Pricing should be explicable in those terms or it is not really explicable at all.
Questions
Should I get multiple proposals?
Yes, and expect them to differ enormously. Where they agree is usually where your real problem is, and that alone is worth the exercise.
What about the free audit?
Frequently a crawler report with the branding changed, listing hundreds of issues with no priority attached. A useful audit tells you which three things matter and why. Volume of findings is not a measure of quality.
How long should I commit for?
Three months minimum to see anything meaningful, and be wary of anything over twelve. A provider confident in their work does not need to lock you in for two years.
