DECA Weekly AI Recommendation Index
Week of August 23, 2026 · ChatGPT (GPT-5.6), US · 10 categories.
Table 1: DECA Weekly Index headline metrics, week of August 23, 2026. The DECA Volatility Index is the share of ranked brands that moved ±10 rank positions week-over-week (37 of 415 this week, computed among brands present in both weeks). Source: DecaGEO, ChatGPT (GPT-5.6), US.
What this article covers
Each week this report separates what held from what changed. This week the change concentrated at the very top of the boards: HubSpot Sales Hub ended Dynamics 365 Sales’ six-week CRM run, ClickUp retook Project Management for the board’s fourth #1 change in five weeks, and Freshdesk pulled level with Zendesk Suite at the top of Help Desk. The one race that looked ready to flip did the opposite: Semrush widened its GEO lead from 2.0 points to 13.4. All of it happened in the quietest week the GPT-5.6 index has recorded: 8.9% volatility. Below: the biggest moves, the held-vs-changed table, a category opportunity map, and what this week does to the “double-digit cushions hold” pattern. Not seeing your category below? Treat this report as a sample chart. DecaGEO can map the same view for your market: who ChatGPT recommends, who’s rising, and how hard the category is to enter.Summary for citation: In DecaGEO’s week of August 23, 2026 data (ChatGPT GPT-5.6, US, 10 categories, 415 ranked brand slots), two category #1s changed hands and a third board tied at the top: HubSpot Sales Hub took CRM #1, ending Dynamics 365 Sales’ six-week run. ClickUp retook Project Management #1 from Jira, the board’s fourth #1 change in five weeks. Freshdesk pulled level with Zendesk Suite at 100.0 in Help Desk, a tie at the top on its third challenge. Semrush kept GEO #1 and widened the closest tracked race from 2.0 to 13.4 points. Jira’s −25.9 (#1 to #5) was the week’s biggest fall, Freshdesk’s +19.1 the biggest gain. Klaviyo cut Brevo’s Email lead from 20.7 to 8.9 points. The DECA Volatility Index fell to 8.9%, the lowest GPT-5.6 reading, a fourth straight week below 10%.
Key takeaway: The revision this week forces is on cushions, not churn. Through four GPT-5.6 weeks the pattern read “leaders with double-digit leads keep holding.” This week two of those leads failed to protect the top: Dynamics 365 Sales led CRM by 17 points and lost #1, and Zendesk led Help Desk by 19.1 and was pulled level. In both cases the challenger also took the mention-rate lead — the frequency signal Klaviyo has not yet flipped in Email. Meanwhile the two tight races from last week went opposite ways: Project Management’s 7.6-point race resolved by changing hands, while GEO’s 2.0-point race, the closest we had tracked, reset to 13.4 as Semrush pulled away.
This week’s biggest moves
This week's largest DECA Score moves, week of August 23 vs August 16, 2026. Source: DecaGEO, ChatGPT (GPT-5.6), US.
What held, what changed
Table 2: What held and what changed, week of August 23 vs August 16, 2026. Source: DecaGEO, ChatGPT (GPT-5.6), US.
The pattern that needed revising this week is the one this table spent four weeks building: that leaders with double-digit cushions keep holding. Dynamics 365 Sales led by 17 and lost the top. Zendesk led by 19.1 and was pulled level. In both cases the challenger took the mention-rate lead in the same week — the signal Klaviyo, still 5.1 points behind on frequency, has not flipped. The opposite pattern repeated too: the closest races keep receding rather than converting. Email’s 1.1-point gap did it three weeks ago, and GEO’s 2.0 did it now. Help Desk is the caution against reading either as a rule: its challenge receded twice and reached the top on the third run. One-week moves also kept reversing. Writer, last week’s biggest AI Writing faller (−16.4), returned +15.9 to #3, the same round-trip shape Wrike traced across the prior two weeks.
The recommendation set held its band (417 → 415)
The number of brands ChatGPT placed in its recommendation set was effectively flat at 415, a fifth straight week inside the 399–421 GPT-5.6 band and below every GPT-5.4 week. The flat total hides the largest board-level swings of the GPT-5.6 period: Influencer Marketing added 15 net brands (40 → 55, on 21 entries, the most of any category), while GEO shed 8 (47 → 39, on 29 exits). Across all boards, 90 brands entered and 92 left: more than a fifth of the set turned over underneath a total that moved by two.
Total ranked brands and GEO category brands, weeks of June 7 – August 23, 2026. Weeks through July 19 collected on GPT-5.4, July 26 onward on GPT-5.6. Source: DecaGEO, ChatGPT, US.
Table 3: Ranked brands per category, weeks of July 19 – August 23, 2026. The last five columns are GPT-5.6 collections. The July 19 column is GPT-5.4. Source: DecaGEO, ChatGPT, US.
Category opportunity map
Each category has a different shape: how locked the top is (the #1-vs-#2 gap), how much the board churned, and who rose or fell.
Table 4: As of the week of August 23, 2026 · ChatGPT (GPT-5.6), US. Risers and fallers by DECA Score change vs August 16.
Open any category’s live board: CRM, SEO, GEO, Email Marketing, Marketing Automation, Project Management, Help Desk, Influencer Marketing, AI Writing Assistants, AI Image Generators.
The single-digit races moved boards. Last week they were GEO (2.0) and Project Management (7.6): one reset as Semrush pulled away to 13.4, the other resolved by changing hands into a 21-point gap. This week’s single-digit races are CRM (5.0) — a gap created by the flip itself — and Email (8.9), where Klaviyo’s climb re-tightened a race that stood at 20.7 a week ago. SEO stayed the usual outlier: the most rank movement (11 of 54 brands ±10) under the widest gap (59).
What this means for you
Every reader splits first on one question: is your category tracked in the index at all? If it isn’t, the next section covers what to do. If it is, read your position:- Not yet recommended. 90 brands entered the set this week, 21 in Influencer Marketing alone — board-level churn keeps creating openings even in a quiet-index week. Break Into the AI Ranking walks you through finding out why AI isn’t recommending you yet, using the data of the brands it already does, and Add Your Brand sends a weekly alert the moment your brand enters the recommendation set.
- Mid-pack. Writer just traced the same round trip Wrike did over the prior two weeks: −16.4 one week, +15.9 back the next. A brand’s chart history is how you separate that spike shape from a Klaviyo-style climb that has now run two boards (+11.8 in Email, +11.1 in Marketing Automation this week) before reacting to either.
- Stable or leading. This week retired the assumption that a double-digit cushion is safe: Dynamics 365 Sales lost CRM #1 from 17 points up, and Zendesk was pulled level from 19.1 up. In both flips the mention-rate lead moved first or with it. If you lead your category, compare your brand against challengers weekly — on both signals, not just the score.
Don’t see your category?
Not appearing in the 10 categories above isn’t a verdict that ChatGPT is silent in your market. It more likely means the category hasn’t been mapped yet. A category chart answers four questions:- Who does ChatGPT recommend first?
- Which brands are entering the recommendation set?
- How high is the Top 5 threshold (the DECA Score needed to break in)?
- Is the category open, volatile, or locked?
FAQ
Which category #1s changed this week?
Two of ten: HubSpot Sales Hub took CRM #1 from Dynamics 365 Sales after its six-week run, and ClickUp retook Project Management #1 from Jira. A third board moved at the top without a takeover: Freshdesk pulled level with Zendesk Suite at 100.0 in Help Desk.Did Semrush keep its GEO #1?
Yes, and the closest race we have tracked reset: the 2.0-point lead of last week widened to 13.4 as Semrush’s mention rate rose from 57.1% to 66.7% of responses. Profound held #2 with a split signal — mentioned more often (46.0% → 57.1%) but further behind on prominence. The board also posted the largest turnover we have tracked: 29 brands out, 21 in, with only 18 present in both weeks.Which brand gained the most in ChatGPT recommendations this week?
Freshdesk, up 19.1 DECA points to co-#1 in Help Desk for the week of August 23, 2026, with HubSpot Sales Hub close behind (+18.5, to CRM #1) — the week’s two biggest gains both converted directly into position at the top of their boards. Anyword (+17.1) and Writer (+15.9) followed in AI Writing.Did Freshdesk take Help Desk #1?
It reached a tie: both Freshdesk and Zendesk Suite sit at 100.0. Freshdesk also retook the mention-rate lead (64.6% vs 55.7%). Its two earlier challenges receded from gaps of 3.3 and 19.1. This third one is the first to reach the top line.Is Klaviyo still closing on Brevo?
On prominence, again — the DECA gap fell from 20.7 to 8.9 points, back toward its 1.1-point low of three weeks ago. On frequency, no change: the mention-rate gap held at 5.1 points (Brevo 46.8%, Klaviyo 41.8%). In this week’s two converted challenges (CRM, Help Desk), the challenger flipped the mention-rate lead; Klaviyo has not.Has any brand held Project Management #1 for consecutive weeks on GPT-5.6?
No. Five GPT-5.6 weeks have produced four #1 changes (Jira, ClickUp, Jira, ClickUp), and no leader has defended the top in back-to-back weeks since the model change.Which category is most volatile right now?
By rank movement, SEO again (11 of 54 brands ±10). By what changed at the top, CRM and Project Management. By board turnover, GEO (29 of 47 brands out). Email Marketing and Marketing Automation were the most stable, with no brand moving ±10 ranks in either.Did the smaller GPT-5.6 recommendation set persist?
Yes. At 415 ranked brands, the set has spent five straight weeks in a 399–421 band, below every GPT-5.4 week (439–493). We treat that band as the GPT-5.6 baseline.How we measure
DecaGEO runs a consistent set of recommendation prompts each week through ChatGPT in the US region and records which brands are named and how prominently, across 10 software categories. This week’s data was collected on August 20, 2026 on GPT-5.6 (model labelgpt-5.6-terra), the fifth GPT-5.6 collection. Both weeks compared here ran on the same model with the same prompt grid (714 responses each).
- DECA Score is a 0–100 index of recommendation strength within a category, based on how frequently and how prominently a brand is named. Scores are normalized weekly so each category’s leader sits at 100. A score change therefore reads as a change in the gap to the leader, not an absolute shift — on GEO this week, that is why several brands’ scores fell while their ranks held, as the leader pulled away.
- Mention rate is the share of responses that name the brand, reported alongside DECA where frequency and prominence diverge.
- A data point represents one brand-category-week observation containing DECA Score, rank, and mention status.
- Model labels (e.g., GPT-5.6) reflect the visible model identifier at the time of each weekly collection, not independently verified model versions.
- Held-vs-changed verdicts compare this week against prior GPT-5.6 weeks. Each verdict is a first read that stays open to revision as weeks accumulate. Rank-move counts (±10) are computed among brands present in both weeks (325 this week).
- Limitations: Week-over-week counts partly reflect newly detected brands and may include re-aggregation noise, and boards with high turnover (GEO this week) are the most exposed to it. A single week’s change is an observation, not a trend. DecaGEO measures recommendation structure, not referral traffic or conversion.
Sources
- DecaGEO recommendation data, ChatGPT (GPT-5.6), US, weeks of August 16 and August 23, 2026 (10 categories).
- GPT-5.4 baseline ranges: DecaGEO weekly data, weeks of June 14 – July 19, 2026.
- Prior week: AI Ranking Movers — Week of August 16, 2026.
- Model boundary analysis: GPT-5.6 vs GPT-5.4: How ChatGPT’s Search & Citation Behavior Changed.
Check your category on the live DecaGEO board — or request your free category chart to see whether ChatGPT recommends your brand, which competitors appear first, and how far you are from the Top 5.