Foundations
Part of When to build keyword research foundations for an England audience
In-house team or agency for keyword research business models?
Compare in-house and agency keyword research business models in England, covering cost, control, demand signals and the right choice for 2027 planning.
What to take away
- Run keyword research in-house if you publish weekly, hold first-party sales data and can fund a specialist.
- Buy an agency if you need a fast start, niche category experience or cover during a hiring gap.
- Neither model is automatically cheaper. In-house costs are a salary plus software, and agency costs are a retainer or project fee.
- Control follows whoever owns the keyword map and its history, which is usually the in-house team.
- For 2027, favour the model that can read AI-generated answers and zero-click results, not just blue-link rankings.
What the two models actually are
In-house means a salaried marketer or content lead owns the keyword map. They see sales data, returns and customer emails, so they can answer a demand spike the same day.
An agency delivers research for a retainer or a project fee. You buy analysts, paid platforms and category experience, but the method usually stays with the supplier.
Demand differs by city and by season. Our guide to keyword research demand signals in England explains how to read those patterns before you commit to either model.
In-house
- Cost shape
- Salary, employer costs, tools, training
- Control
- You own the keyword map and its history
- Speed
- Same-day reaction to demand spikes
- Tooling
- You buy and learn the platforms
- Risk
- Single point of failure if one person leaves
Agency
- Cost shape
- Retainer or project fee, cancellable at renewal
- Control
- Supplier holds the method and tooling
- Speed
- Faster start, slower per-request turnaround
- Tooling
- Paid platforms included in the fee
- Risk
- Switching supplier can mean starting again
In-house vs agency cost and control
In-house
- Cost
- Predictable salary
- Example
- £32,000 salary
- Overheads
- Employer overheads
- Control
- Owns keyword map
- Switching
- Keeps history
Agency
- Cost
- Retainer or project fee
- Example
- £1,500 monthly retainer
- Overheads
- No employer overheads
- Control
- Holds tooling and method
- Switching
- May start again
What each model costs
Both routes carry labour and software, so compare them on the same 12-month scope. Figures here are illustrative, not market averages.
Costs split three ways. In-house brings a salary, employer costs, software and training, and the spend repeats in quiet months. Agency brings a retainer or project fee with a defined scope and a renewal date.
Software is common to both. Google Keyword Planner and Search Console are free. Paid platforms such as Semrush and Ahrefs charge by seat and usage limits.
The familiar comparison for a Leeds firm is a £32,000 salary against a £1,500 monthly retainer. The retainer equals £18,000 a year, leaving a £14,000 cash gap before employer costs. It buys fewer hours, but no employer overheads.
A keyword research budget template helps you size the software line before you gather quotes.
Checklist: which model fits
- You publish at least weekly and can fund a specialist rolein-house is the better base.
- You hold first-party sales, returns and support datain-house will use it faster.
- You need a quick start, a new category or maternity coveran agency stands up sooner.
- You have one marketer and a modest budget: run a hybrid, keeping strategy in-house.
A decision framework for 2027
- Count the pieces you will publish over the next 12 months. Below about 20, an agency or hybrid usually fits better.
- Total the in-house costsalary, employer costs and two paid keyword platforms.
- Get two agency quotes for the same scope, one on retainer and one on a fixed project fee.
- Score control and handover risk, because losing the keyword map costs more than a fee difference.
- Review the choice every 12 months, or sooner if publishing volume or target regions change.
How AI search changes the trade-off
AI overviews and zero-click answers reduce the value of position checks alone. In-house teams see query data first through the Search Console performance report, which lists the real queries and click-through rates behind a page.
Agency tooling offers a different view. A metric such as Ahrefs keyword difficulty scores how hard a term looks, but it measures links and authority rather than whether an AI answer removes the click.
Market context still matters. The ONS business, industry and trade hub carries UK sector statistics you can mine for content angles.
Whoever owns your research should also own AI visibility checks in 2027. Our round-up of keyword research 2027 trends covers the sources worth budgeting for.
Common questions
Is an agency always more expensive?
No. A small retainer can cost less than a full salary plus tools, but you get fewer hours and less day-to-day control.
Can a small team run research in-house?
Yes, if one person owns the process and reviews it monthly. Start with a narrow set of terms tied to your best-selling products.
How often should we review the model?
Annually, or when publishing volume, budget or target regions change. A model that suited a startup often stops fitting a firm of 40 staff.
Does location within England change the choice?
Yes. Phrasing and competition differ between London, Birmingham and smaller cities, so test regional terms before assuming one national pattern.
