Service page or blog post: which one should target the keyword?

Picture a hypothetical plumber who publishes a 1,800-word post called “Everything you need to know about drain cleaning” and waits for the phone to ring. It does not, because the people typing “drain cleaning” plus a city name are not looking for an essay. They want a van.

How do you read the results for a keyword?

Search the keyword, label every result on the first page by type, and build the type that wins. Google has already decided what the query deserves.

Your job is to match it, then be better than what’s there. The routine below is ours; Google doesn’t publish one.

1. Open a private windowSet your service area so history and location do not skew results.
2. Label the first ten resultsMark each as a service page, guide, cost page, forum and so on.
3. Note the extrasLook for a map pack, ads, People also ask and video.
4. Tally and decideCount the labels and read the decision table.
5. Record the owner URLWrite the decision in your keyword map.

1. Open a private window and set your service area, so history and location do not skew the results.
2. List the first ten organic results and label each: service page, location page, guide, comparison, cost page, directory, forum, video.
3. Note the extras on the page: map pack, ads, “People also ask”, video carousel, discussion threads.
4. Tally the labels and read the table in the next section.
5. Write the decision and the owning URL in your keyword map, so nobody creates a competing page next quarter.

What two labeling rules save arguments?

A directory listing with a “call now” button counts as buy intent. A page under /blog/ that is built to sell counts as a service page, because the page decides, not the folder name.

Intent shifts with the modifier (“cost”, “near me”, “vs”), with the city, and over time. So we re-read a site’s highest-value keywords every few months.

What thresholds do we use to choose?

These thresholds are ours, not Google’s, and they work as a tiebreaker rather than a law. When the count is close, trust the extras from step three.

How does the table work on real keywords?

Here are three hypothetical tallies, so you can see how the table gets used. None is a measured result.

• “Water heater repair” plus a city: six service or location pages, two directories, one guide, one video, and a map pack. Build a service page.
• “Tankless vs tank water heater”: seven articles, two videos, one service page, no map pack. Write a comparison post and link it to the service page.
• “Furnace not heating”: three service pages, four guides, two forum threads, one video, and a map pack. No type reaches six, so build a service page for the repair and a troubleshooting post, each with its own query.

Why does the map pack get extra weight?

The map pack deserves its weight. Google says local results are mainly based on relevance, distance, and popularity, which its page explains as prominence.

So in our reading, a query that triggers a map pack is about a nearby business. A blog post rarely competes there, and the profile does most of the work.

Does research back this up?

Intent mismatch also shows up in research. Ahrefs lists it as one of three reasons pages get no search traffic in its study of about 14 billion pages. It’s a vendor study with caveats, but the lesson holds: a good page of the wrong type is still the wrong page.

What do local service queries look like in practice?

They fall into five shapes, and each shape tends to want a different page. The examples are hypothetical classifications, not measured results, so check your own city.

One service therefore produces at least four page jobs, which is why “service page or blog post” is rarely a question about a whole topic. It’s a question about each keyword. A keyword map for the same hypothetical company might look like this:

water heater repair phoenix      -> /water-heater-repair/phoenix/   (service page, owner)
emergency plumber near me        -> /emergency-plumber/             (service page + GBP link)
water heater replacement cost    -> /blog/water-heater-replacement-cost/
tankless vs tank water heater    -> /blog/tankless-vs-tank/
water heater leaking from bottom -> /blog/water-heater-leaking/     (links to repair page)

Every query has exactly one owner URL. If you catch yourself writing two URLs for one query, resolve it before anything is drafted.

For the full build behind the first row, our service page SEO guide covers structure and copy. For a map-pack query, the local SEO guide covers the profile side, and what search intent means for page type goes deeper on reading results.

What does Google say about multiple pages for similar queries?

Google generally limits a site to two listings in the top results and removes near-identical pages. Its spam policies also describe doorway abuse: pages made to rank for similar queries that funnel users to one place. Two competing pages rarely help you, but nothing we read says they are penalized.

What do the documents say?

Google’s ranking systems guide was last updated 10 December 2025 on the page. It says “we generally won’t show more than two web page listings from the same site in our top results.” It also lists a deduplication system that shows only the most relevant of very similar results.

The spam policies, last updated 28 August 2026, define doorway abuse as “when sites or pages are created to rank for specific, similar search queries.” One listed example is “creating substantially similar pages that are closer to search results than a clearly defined, browseable hierarchy.”

What does that mean for a second page?

So the benefit of a second page is small, and the risk grows when the pages are near copies. The documents we read do not use the term “cannibalization”, so treat it as industry shorthand rather than Google vocabulary.

Our post on location pages without doorways shows where that line sits for city pages. The glossary entry on keyword cannibalization gives the short definition.

How do you check whether two of your pages clash?

Open Search Console, filter to the query, and see which of your URLs Google shows for it. If two URLs split the impressions, you have a clash. If one URL holds them all, you don’t, whatever its type.

Here are the steps we use in the Performance report of Search Console, in order:

1. Open Performance, then Search results, and set the date range to the last three months or longer.
2. Add a Query filter for the keyword you are checking (“Queries containing” catches variants).
3. Switch to the Pages tab and read which URLs have impressions and clicks.
4. Click each URL, then switch to the Queries tab to see what else it ranks for.
5. Decide which URL owns the query, using the thresholds and the job each page does.

What do made-up numbers show?

Take made-up numbers for “drain cleaning”. The service page shows 620 impressions and 11 clicks, and a blog post shows 410 impressions and 9 clicks. Together that is 1,030 impressions and 20 clicks.

• Service page620
• Blog post410

Source: Hypothetical example

View the data as a table

Item	Value
Service page	620
Blog post	410

The service page holds about 60% of impressions (620 of 1,030) and the post about 40%. Click-through rates are 1.8% (11 of 620) and 2.2% (9 of 410). The post converts impressions to clicks slightly better, which is a reason to look at calls before deleting anything.

• Service page1.8%
• Blog post2.2%

Source: Hypothetical example

View the data as a table

Item	Value
Service page	1.8%
Blog post	2.2%

Why read the Pages tab, not the average?

Clicks count how often users went from Google results to your site, and impressions count how often your site appeared. Average position reflects the topmost result from your site, so it can hide a swap between two URLs. That’s why we read the Pages tab, not the average.

Start with your ten most valuable keywords. Our Google Search Console guide explains the rest of the report.

What do the fixes look like, before and after?

The fix depends on which type the results want and which type you built. Each example below is hypothetical, written to show the rewrite, and is not a client result.

Example 1: a post built for a service keyword

A hypothetical HVAC company has a post titled “Furnace Repair: Everything You Need to Know”. Results for “furnace repair” in its city show service pages and a map pack.

Example 2: a service page built for a question

A hypothetical roofer has a service page that tries to answer “How much does a roof replacement cost?” The results show cost guides and ranges.

Example 3: two pages on one query

A hypothetical plumber has a service page and a post for “drain cleaning”, with near-identical introductions.

Our redirects guide covers the mechanics, and the redirect mapping guide helps when there are many. After the merge, load the old URL and confirm it reaches the new one in a single hop.

Can one page do both jobs?

Only in small doses. A service page can carry a short FAQ and a cost summary, because those answer questions a buyer has at the point of decision. It can’t carry a full explainer without hurting the sale.

The test we run on each section: would it make a good standalone answer to a different query? Then give it its own page and link to it. If it only helps a buyer decide, it stays on the service page.

What can a service page safely hold?

A service page can safely hold these quick answers for buyers:

• A three-to-five question FAQ on timing, coverage, and what a visit involves
• A cost summary of one or two sentences, with a link to the full cost post
• A booking path visible without scrolling

A post, in turn, can carry a call to action and a link to the service page. It should not try to be the service page.

What does Google ask about value?

Google’s helpful-content guidance, last updated 5 October 2026, asks whether content provides “substantial value when compared to other pages in search results” and whether it is “primarily made to attract visits from search engines.” A post that exists only to sell fails the first test and invites the second.

Links hold a pair together. Google says good anchor text is “descriptive, reasonably concise, and relevant to the page that it’s on and to the page it links to.” Our internal linking guide shows how we connect the two.

What editorial rules do we apply before a page ships?

Six rules, applied to our own pages and to client pages alike. They’re our practice, not an industry standard, and each exists because skipping it made a page harder to judge later.

1. One query, one owner. The brief names the URL that owns the keyword. If a second page is needed, it gets a different query.
2. The title leads with the job. A service page title names the service and the place. A post title names the question.
3. The first 100 words do the page’s job. On a service page that means service, area, and next step. On a post it means the answer.
4. Every post links to its service page once, in context. The anchor describes the service, never “click here”.
5. Every service page links down to its supporting posts. Cost, comparison, and symptom posts are reachable in one click.
6. A review date goes in the brief. We note when the results for the keyword get re-read.

What does rule 3 look like?

Here is rule 3 on a hypothetical symptom post.

Rule 6 is the one teams drop first. The brief format that carries all six is in our content brief guide.

What mistakes do we see most often?

Most mistakes come from deciding by habit instead of by the results. This list is our judgment from audits, not a count.

Our post on what a service business should publish ranks page types by their link to revenue. For the blogging side of local rankings, see whether blogging helps local SEO for home-service companies.

What does the data not show?

• It does not tell you what your keyword will return, because that varies by city and over time. The thresholds here are our rule of thumb, and Google does not say how many service pages in the results make a query transactional.
• We also have no study comparing conversion rates of service pages and posts, so we cite none. The local examples are hypothetical classifications, and the numbers in the worked examples are invented to show arithmetic.
• The documents we read describe site diversity and deduplication, not how Google treats a particular pair of your pages. The only real evidence is Search Console data for your own site.

Our content audit guide shows how to gather it site-wide.

Frequently asked questions

Next step

Take your five highest-value keywords, run the five-step read, and write down the page type each one wants. Our SEO content brief generator turns each decision into a brief, and our content SEO service works from the same rule. Then Get Your Organic Growth Plan.

Sources

• Google Search Central: A guide to Google Search ranking systems, last updated 10 Dec 2025 on the page, accessed 22 Oct 2026
• Google Search Central: Spam policies for Google web search, last updated 28 Aug 2026 on the page, accessed 22 Oct 2026
• Google Search Central: Creating helpful, reliable, people-first content, last updated 5 Oct 2026 on the page, accessed 22 Oct 2026
• Google Search Central: Link best practices, last updated 10 Dec 2025 on the page, accessed 22 Oct 2026
• Google Search Central: SEO Starter Guide, last updated 10 Dec 2025 on the page, accessed 23 October 2026
• Google Business Profile Help: Improve your local ranking on Google, accessed 22 Oct 2026
• Google Search Console Help: Performance report, accessed 22 Oct 2026
• Ahrefs: 96.55% of Content Gets No Traffic From Google, published 1 Dec 2023, accessed 22 Oct 2026