Test log
AiPPT test data
Raw results for the AiPPT review. The verdict is in the review.
- Tested
- 6 November 2026
- Tester
- Miles Foster
- Account
- Free demo account, signed up by RankDude, nothing paid to the brand
- Tests
- 10
- Screenshots
- 8
- Method
- Published methodology
- Environment
- Google Chrome on desktop
Results
| ID | Test | Key measurement | Score |
|---|---|---|---|
| T01 | Use of a Supplied Document | – | n/a |
| T02 | Brand & Theme Control | – | n/a |
| T03 | Export Fidelity | – | n/a |
| T04 | Structure from a Prompt | Structure points met: 11 of 14 | 3/5 |
| T05 | Slide Design | Layout faults: 3 | 4/5 |
| T06 | Content Accuracy | Figures checked: all slides | 4/5 |
| T07 | Editing Control | Edits that worked: 7 of 12 | 3/5 |
| T08 | Charts & Images | Charts correct: 2 of 6 asked for | 2/5 |
| T09 | Speed | Classic: outline to finished deck: about 40 seconds | 5/5 |
| T10 | Pricing & Value | Cheapest plan with export, billed yearly: $9 a month | 3/5 |
| Overall (in the review) | 3.8/5 | ||
Use of a Supplied Document
n/aNot Tested- Method
- A pitch deck from a six-page document, checked against 15 key points listed beforehand.
- Remark
- Not graded; left out of the average.
Raw notes
A pitch deck from a six-page document, checked against 15 key points listed beforehand.
Not run: the paperclip in the create box opened the upgrade box on the free plan. The plan table lists "Import from local files" as not included in Free and included in Plus and Pro.
- Needs a paid plan.
Brand & Theme Control
n/aNot Tested- Method
- Add the logo, two hex colours and a font, import a .pptx template and check every slide.
- Remark
- Not graded; left out of the average.
Raw notes
Add the logo, two hex colours and a font, import a .pptx template and check every slide.
Not run: I found no brand kit on the free plan, and the .pptx template import sits under the paid local file import. A logo can be dropped on a slide by hand as an image, which worked, but a deck that applies the kit to every slide needs credits I no longer had.
- Needs credits beyond the free 140 and a paid plan.
Export Fidelity
n/aNot Tested- Method
- Export each deck to PowerPoint, Google Slides and PDF and compare slide by slide.
- Remark
- Not graded; left out of the average.
Raw notes
Export each deck to PowerPoint, Google Slides and PDF and compare slide by slide.
Not run: choosing Download in the editor opened the upgrade box. The plan table lists "PPT/PDF/Image export" as not included in Free.
- Needs a paid plan.
Structure from a Prompt
3/5Tested- Method
- The brief for Northwind Cycle Co.
- Structure points met
- 11 of 14
- Slides built
- 10 of 10 (Classic), 5 of 10 (Agent)
- Remark
- Scale 5: 20 or 21 of 21 points; 4: 17 to 19; 3: 13 to 16. With two drafts the scale is cut to 14 points, where 4 starts at 11.3, so 11 of 14 is a 3.
Raw notes
The brief for Northwind Cycle Co. asked for ten slides: a title slide, an agenda, a bar chart of revenue by region, a line chart of monthly revenue, a KPI slide, risks, next steps and a closing ask. Both first drafts were checked against the seven-point list in the methodology.
The Classic draft met six of seven points. It had ten slides, a title slide, a logical order from Q3 results to next steps, one idea per slide, a closing slide and no repeats. It had no agenda slide, although the brief asked for one.
The Agent draft met five of seven. It had a title slide, an agenda of eight items, a logical order and one idea per slide, and no repeats. It stopped at slide 5, so it had five of the ten slides and no closing ask.
- One brief, built twice. The water cycle and pitch decks were not generated, so 7 of the 21 points were not checked.
Slide Design
4/5Tested- Method
- The ten Classic slides and the five Agent slides were captured at 2x and checked for text overflowing its box, overlaps, misalignment, mixed fonts, walls of text and unrelated images.
- Layout faults
- 3
- Faults per 10 slides
- 2.0
- Remark
- Scale 4: 1 to 2.9 faults per 10 slides.
Raw notes
The ten Classic slides and the five Agent slides were captured at 2x and checked for text overflowing its box, overlaps, misalignment, mixed fonts, walls of text and unrelated images. Faults were counted and expressed per ten slides.
Three faults in 15 slides. On the Classic cover and again on the closing slide a phone image with the words "DIGITAL BANKING" sits behind a cycling company's title. On slide 3 a black block pokes out behind one of the two stock photos.
The Agent slides had no faults: clean spacing, one font family and two colours. I did not count the "YOUR LOGO" placeholder or the template line "Here is where your presentation begins" on the Classic cover as layout faults; they are left-over template text. The editor also floats a "How would you rate this presentation?" box over the corner of the canvas.
- Faults were counted by eye at one window size. Text contrast was not measured with a checker. Text on Classic slides 3, 7 and 8 is small.
Content Accuracy
4/5Tested- Method
- Every figure on the slides was compared with the table in the brief: three months of revenue, orders, new customers and support tickets, revenue by region for Q2 and Q3, the totals, margin and satisfaction.
- Figures checked
- all slides
- Wrong or invented figures
- 0
- Minor errors
- 1
- Remark
- Scale 4: one minor error.
Raw notes
Every figure on the slides was compared with the table in the brief: three months of revenue, orders, new customers and support tickets, revenue by region for Q2 and Q3, the totals, margin and satisfaction.
Every figure on the five Agent slides matched the table: $1,368k against $1,214k, up 12.7 per cent; July, August and September at $412k, $455k and $501k with 3,180, 3,490 and 3,870 orders; North, South, East and West at 388 and 452, 301 and 365, 274 and 331, 251 and 220; 41 per cent margin and 4.6 out of 5. It called September the best month and West the only region to fall, and both are true.
The Classic deck carried none of the figures. Its slides say things like "Analysis of total revenue generated in Q3 compared to Q2" with no number, and the cover and closing slide kept template text. I counted that as one minor error.
- Only the quarterly review was checked. The water cycle facts were not checked because that deck was not generated.
Editing Control
3/5Tested- Method
- Twelve fixed edits on the Classic quarterly review deck.
- Edits that worked
- 7 of 12
- Credits for a theme change
- 5
- Remark
- Scale 3: 7 to 9.
Raw notes
Twelve fixed edits on the Classic quarterly review deck. An edit counts when it did what was asked with a normal control and no workaround.
Seven of twelve worked: a slide reordered by drag, a blank slide inserted after the selected slide, a native chart inserted and its data edited in a grid, an image replaced from the deck gallery and from my own drive, a text box moved by drag, one theme applied to the whole deck, and three undo steps with the keyboard.
Five did not: I found no control to change a slide's layout, to regenerate one slide, to shorten a paragraph with AI, to add speaker notes or to leave a comment for a collaborator. The share box has link access, a password and analytics only.
Dragging a slide in the strip moved it one place higher than I aimed. Applying a theme cost 5 credits. The chart edit worked on a chart I inserted: the Classic deck's own slides held no chart.
- The Agent deck has an Edit page button on each slide; I did not run the twelve edits on it because its credits were gone.
Charts & Images
2/5Tested- Method
- The brief asked for a bar chart, a line chart and a KPI panel.
- Charts correct
- 2 of 6 asked for
- Charts in the Classic draft
- 0 of 3
- Remark
- Scale 2: two or more charts wrong. Across both drafts two of the six requested charts were right, one was not reached and three were photos or text.
Raw notes
The brief asked for a bar chart, a line chart and a KPI panel. Each value was checked, and the images in both drafts were checked for relevance and source.
The Agent deck's bar chart carried all eight values and the line chart carried the three months, each with a source line reading "Source: Northwind Cycle Co." and the figures. Its KPI slide was never built because the task paused.
The Classic deck had no chart at all. On slide 6, titled "Revenue by Region Analysis", a stock photo of coloured bars sits where the bar chart should be, and slide 3 does the same for the line chart. The KPI slide lists headings with no figures.
A native chart inserted in the editor has a data grid and editable values. Whether the Agent's charts can be edited as data I did not test. Stock photos carry no source or licence on the slide.
- Two drafts of one brief, so the grade mixes a draft that works and one that does not. The Agent charts were not clicked to check for data behind them.
Speed
5/5Tested- Method
- The Classic brief was typed with Pages set to 5-10 and sent.
- Classic: outline to finished deck
- about 40 seconds
- Agent: five slides after Skip all
- about 1.5 minutes
- Credits: Classic deck, Agent run
- 20 and 104
- Remark
- Scale 5: median under 1 minute. Only the Classic run finished, so it sets the median; the stalled Agent run is reported here and not counted.
Raw notes
The Classic brief was typed with Pages set to 5-10 and sent. The clock ran from confirming the outline, when the Generate button shows its price, to the last slide finished and editable. The Agent run was timed from my Skip all click to the time its history list shows.
The Classic deck was finished about 40 seconds after I confirmed the outline: the editor showed 0 per cent at 5 seconds, 50 per cent at 10 and 90 per cent at 15, then the finished deck at about 40 seconds. The outline step shows the price before you confirm it: "Generate 20" credits.
The Agent run took the full brief and spent 33 credits before the first slide, reading its skills and asking two questions: what the other two slides should cover, and the visual style. After I chose Skip all it wrote five slides in about a minute and a half, for 71 more credits, and then paused with "Your remaining credits are insufficient."
- One finished deck, so the median is that one run. The Classic outline time was not logged on its own.
Pricing & Value
3/5Tested- Method
- Plans were read in the upgrade box inside the editor, in the monthly and the yearly view, and in its Compare Plans table.
- Cheapest plan with export, billed yearly
- $9 a month
- Same plan, monthly
- $12
- Remark
- Scale 5: free, or under $10 a month, for the cheapest plan that exports .pptx; 2: unclear credit costs. I set 3 between the two.
Raw notes
Plans were read in the upgrade box inside the editor, in the monthly and the yearly view, and in its Compare Plans table.
The cheapest plan that exports .pptx is Plus at $12 a month, or $9 a month billed yearly ($108 a year). Pro is $16 a month, or $11 billed yearly. Plus gives 600 credits a month and 20 slides per AI generation; Pro gives 1,200 and 50.
The free plan gives 140 credits once and blocks local file import and every export. The price band alone would earn a 5, but credit use is hard to predict: Classic shows its price first (20 for a deck), while the Agent run showed none and drew 104 credits for five slides. That pulls the grade down to 3.
- No plan was bought, so the watermark question on a paid export was not tested. The Lifetime tab was not read.
Other features checked
Coverage
Tested
- The quarterly review brief typed into Classic (cut to the 500 characters the box allows) and into Agent mode in full
- Seven structure points, layout faults on 15 slides, every figure against the table and the charts checked
- Twelve fixed edits in the Classic editor, with a capture after each
- Upload button, Download menu, plan chooser in both billing views, Terms of Use and Trustpilot read
Not tested or limited
- Only one of the three fixed briefs was graded, built twice. The Agent run took the credit meter from 120 to 16, and its cost is not shown before it starts, so I could not stop at 85 per cent of the allowance
- The water cycle brief was not run: the free plan caps AI generation at 10 slides and the credits were gone
- The pitch deck from a six-page document was not run: the paperclip in the create box opened the upgrade box, because importing local files is a paid feature
- Brand control was not graded: there is no brand kit on the free plan, a .pptx template needs the paid file import, and I had no credits left for a branded deck
- Export fidelity was not graded: Download opened the upgrade box on the free plan
- The Classic box stops at 500 characters, so the Classic deck was built from a shortened version of the brief with every figure and chart request kept; the Agent deck used the brief as written
- Text contrast was read by eye, not measured with a checker. The Agent deck's slides 6 to 10 were never built, so they could not be checked
Still to test
- The water cycle and pitch briefs on a paid month, when credits allow
- The Agent deck's remaining five slides, and whether its charts hold data
- Brand kit and template import on a paid month
- Export fidelity to PowerPoint, Google Slides and PDF on a paid month