DEV Community

Natalie Yevtushyna
Natalie Yevtushyna

Posted on • Originally published at seeklab.io

I Stopped Picking Keywords by Volume. I Use This 3-Part Framework Instead.

#seo #startup #webdev #productivity

Most keyword research workflows have the same flaw: they start and end with volume and difficulty.

A keyword with 10,000 monthly searches and KD 35 looks great on paper. But if the people searching for it are students, wrong-market buyers, or users who already have what they need — the traffic is worthless.

We built the BID Framework after watching too many content calendars optimized for metrics that had nothing to do with business outcomes.

TL;DR: Score every keyword across Business Value (40pts), Intent Match (30pts), and Difficulty Reality (25pts) + a 5pt timing bonus. Anything above 75 goes into production. Below 60 becomes an FAQ, not its own page.

The three questions

Before any keyword gets approved, it has to answer three questions:

B — Business Value
    "If someone reads this article and finds it useful,
     is there a natural next step toward a conversation with us?"

I — Intent Match  
    "Does our page format match what the searcher actually expects to find?"

D — Difficulty Reality
    "Can this specific site realistically compete for this query right now?"
Enter fullscreen mode Exit fullscreen mode

B: Business Value (40 points)

This is the question most keyword tools don't ask at all.

# Business Value scoring guide

def score_business_value(keyword):
    score = 0

    # Does it attract the right buyer?
    if attracts_decision_makers(keyword):
        score += 15

    # Is there a natural next step to a commercial conversation?
    if has_commercial_path(keyword):
        score += 15

    # Does it match your actual service/product category?
    if category_aligned(keyword):
        score += 10

    return score  # max 40
Enter fullscreen mode Exit fullscreen mode

The test: "Best butterfly valve supplier for food processing certification requirements" scores higher on Business Value than "what is a butterfly valve" — even though the second has 10x more search volume.

Volume tells you what's popular. Business Value tells you what will convert.

Low Business Value signals:

  • Searchers are primarily students or researchers
  • The query is answered completely by an AI Overview (zero click potential)
  • The audience is in a market you don't serve
  • The information need has no downstream commercial action

I: Intent Match (30 points)

This is the most commonly skipped check — and the most expensive mistake.

# Intent type mapping

Informational  → Long-form guide, definition page, how-to article
Navigational   → Brand page, login page, specific tool page  
Commercial     → Comparison page, "best X for Y" article, reviews
Transactional  → Product page, pricing page, free trial CTA

# The mistake: publishing the wrong format for the intent

❌ "SEO checklist" → writing a 3,000-word essay about why SEO matters
✅ "SEO checklist" → an actual checklist people can use

❌ "best CRM for B2B" → a definition of what CRM stands for
✅ "best CRM for B2B" → a comparison with verdicts and "Best for:" labels
Enter fullscreen mode Exit fullscreen mode

How to check intent before writing:

# Search the keyword. Look at what's ranking.

Questions to answer:
1. What format is winning? (list, guide, tool, comparison, definition)
2. How long are the ranking pages?
3. What does the searcher expect to find in the first section?
4. Is there a featured snippet? What does it look like?

# If your planned format doesn't match → kill the keyword or change the format
Enter fullscreen mode Exit fullscreen mode

Intent mismatch is why pages with genuinely good content fail to rank. Google can see when the format doesn't match what searchers expect.

D: Difficulty Reality (25 points)

Standard keyword difficulty scores are useful but incomplete. BID's Difficulty Reality check has five components:

1. Competitor authority
   Who's ranking? Large publications you can't touch for 2 years,
   or sites at your domain authority level?

2. Content depth required
   Does winning require original research, expert interviews,
   proprietary data? Or can you produce this well?

3. Topical authority
   Do you have enough related content for Google to treat you
   as authoritative on this topic? Or is this an isolated article?

4. Technical readiness
   Is the site crawlable? Are commercial pages indexed?
   A brilliant article on a site with rendering issues is wasted.

5. Regional opportunity
   Sometimes the global term is too competitive but the same
   query in a specific market or language is wide open.
Enter fullscreen mode Exit fullscreen mode

The 5-factor difficulty score:

difficulty_factors = {
    "competitor_authority": 0-5,   # 5 = all big publications, 1 = similar DA sites
    "content_depth_required": 0-5, # 5 = needs proprietary data, 1 = standard guide
    "topical_authority_gap": 0-5,  # 5 = no related content exists, 1 = strong cluster
    "technical_readiness": 0-5,    # 5 = major issues, 1 = clean site
    "regional_competition": 0-5    # 5 = global term saturated, 1 = clear regional gap
}

# Lower total = better Difficulty Reality score
# Invert and scale to 25 points
Enter fullscreen mode Exit fullscreen mode

The scoring model 

| Dimension        | Max score | Weight |
|------------------|-----------|--------|
| Business Value   | 40 pts    | 40%    |
| Intent Match     | 30 pts    | 30%    |
| Difficulty Reality| 25 pts   | 25%    |
| Timing bonus     | 5 pts     | 5%     |
| TOTAL            | 100 pts   |        |

Decision thresholds:
75+    → Write it. Goes into production calendar.
60-74  → Reconsider. Find a better angle or long-tail version.
<60    → Don't write a standalone page. Use as an FAQ section instead.
Enter fullscreen mode Exit fullscreen mode

A worked example 

Keyword: "SEO roadmap AI gateway startup"

B — Business Value: 38/40
    → Attracts funded AI infrastructure founders (exact SeekLab client)
    → Natural path to free audit → paid engagement
    → Direct category match

I — Intent Match: 28/30  
    → Searcher wants a practical roadmap, not a definition
    → Format: step-by-step guide with timeline ✅
    → Small deduction: query is niche, intent less established

D — Difficulty Reality: 23/25
    → Almost zero competing content exists
    → No large publications ranking
    → SeekLab has topical authority in SEO/GEO
    → Site is technically clean

Timing: +4 (AI gateway category actively growing in 2026)

Total: 93/100 → Write immediately
Enter fullscreen mode Exit fullscreen mode

That article got 5 clicks in day 1. The AI Agent Startup version of the same format got 24 clicks in 3 days.

What to do with sub-60 keywords

Don't discard them — repurpose them:

Sub-60 keyword → FAQ section inside a higher-scoring article
                 Internal link anchor text
                 H3 subsection within a broader guide
                 Briefing/news post (low effort, no ranking expectation)
Enter fullscreen mode Exit fullscreen mode

A keyword that doesn't deserve its own page still deserves to exist somewhere on the site. It adds semantic coverage without diluting topical authority.

The tools BID works on top of

BID isn't a replacement for your existing keyword research tools — it's the decision layer on top of them:

Semrush / Ahrefs    → Volume + standard difficulty (feeds B and D)
Google Search Console → Real query data from your own site (feeds I)
AlsoAsked           → Question-based intent mapping (feeds I)  
Exploding Topics    → Timing signal (feeds the 5pt bonus)
Manual SERP review  → Intent check (feeds I and D)
Enter fullscreen mode Exit fullscreen mode

SeekLab.io — SEO and GEO for brands that need AI systems to recommend them. Free keyword audit at seeklab.io/audit.

Top comments (0)