Comparison
Deconflict is a free, open-source WiFi planner that runs in the browser. So is our free tier. Both are good, and they are built for different jobs: we report in dBm and produce a deliverable you can hand a paying client; they optimise placement and are free forever. Across the 25 rows below we take 13, they take 6, and 6 are even. Here is the whole picture, including the rows we lose.
Every claim checked by hand on deconflict.app or read from its public repository, last verified 2026-08-31.
Deconflict reports
Five relative bands, and a coverage percentage. Good enough to compare two placements against each other, and to see where a room goes dark.
WiFi Heatmap reports
An absolute figure at every point, plus the same plain-English band. The number is what goes in the specification; the band is what the client reads.
This sounds like a detail and is not. Coverage requirements are written in dBm — "−67 dBm or better in all occupied areas" is the standard line in a tender. A relative band cannot answer that, cannot be checked with a free phone app after installation, and cannot be signed off. For planning a house it makes no difference at all. For anything you invoice, it is the difference between a prediction and a deliverable.
13 rows favour us, 6 favour Deconflict, 6 are a genuine tie.
Both tools predict coverage from a floorplan. They report it very differently.
| Dimension | WiFi Heatmap | Deconflict | Edge |
|---|---|---|---|
| Signal units Specifications, SLAs and snag lists are written in dBm. A colour band cannot be quoted back to a client or checked against a phone. | dBm — on the map, on hover, and in the report | Relative bands: Excellent, Good, Fair, Poor, Dead | WiFi Heatmap |
| Propagation model Both are defensible indoor approximations. n = 4 is a reasonable exponent for furnished space; the difference that matters is the units above, not the curve. | COST-231 multi-wall, evaluated as a link budget in dBm | Inverse-quartic path loss (n = 4) with per-material wall loss | Even |
| Wall materials Attenuation rises with frequency. A single per-material figure is optimistic on whichever bands it was not calibrated for. | 7 materials, each tabulated for 2.4, 5 and 6 GHz | 6 materials, thickness-aware, one figure per material | WiFi Heatmap |
| Floor slab attenuation They differentiate slab loss by frequency properly. Credit where it is due — this is the part of their model we would not change. | Per material, per band, summed across every slab crossed | Per material, per band, summed across every slab crossed | Even |
| Antenna gain Gain is part of the link budget. A 6 dBi long-range AP and a 2 dBi wall plate are not the same install. | Modelled per access point; the catalogue carries gain per radio | Not modelled | WiFi Heatmap |
They cover more devices. We cover fewer, on purpose, with the paperwork to back each one.
| Dimension | WiFi Heatmap | Deconflict | Edge |
|---|---|---|---|
| Access point library If breadth is what you need, they have more. Ours is deliberately the gear installers actually mount, and every figure traces back to the manufacturer. | 79 models, enterprise and prosumer, datasheet URL on every entry | About 101 models across 14 vendors, including many consumer routers | Deconflict |
| Bands | 2.4, 5 and 6 GHz | 2.4, 5 and 6 GHz | Even |
| Channel width per AP Width only helps if it reaches the channel plan: at 80 MHz, 36 and 40 are one block, not two channels. Placing two 160 MHz access points on their sample office plan, Deconflict assigned them 36 and 40. | 20 / 40 / 80 / 160 MHz, and the channel plan uses the block each radio occupies | 20 / 40 / 80 / 160 MHz, selectable per access point | WiFi Heatmap |
This is the section where a free tool beats us, and it is worth being blunt about it.
| Dimension | WiFi Heatmap | Deconflict | Edge |
|---|---|---|---|
| Multi-floor Same capability, different price. Theirs also carries floor thickness as a separate input. | Yes — Pro | Yes — free | Deconflict |
| Automatic channel assignment | Yes — Pro. Conflict detection is free | Yes — free | Deconflict |
| Placement optimiser Theirs is free, which is a real advantage. Ours is deterministic — the same plan always returns the same positions, which matters when the result goes into a document someone signs off; theirs seeds from Math.random(), so a re-run can land elsewhere. | Yes — Pro. Deterministic, and scored on your actual wall materials | Yes — free, a three-stage optimiser | Even |
| How many access points do I need The number an installer is actually paid for, and the one a relocating optimiser cannot produce. It also says which limb decided it, so the figure can be argued with rather than just accepted. | Yes — the greater of what coverage needs and what the device load needs | Not available — it relocates the APs you already placed | WiFi Heatmap |
| Wall detection from a floorplan image Different trade-offs. Theirs traces into a raster mask, so it copes with any wall shape. Ours emits real segments you can retype or delete, and merges the two drawn faces of a wall so it is not counted twice — but it only finds axis-aligned walls, and says so. | Yes — traced as editable segments, each with a material | Yes — traced into a wall mask, with OCR to ignore text | Even |
| Room types and device density Both model it. Theirs has more room types; ours states the source and the arithmetic behind each density, and works from a single floor-level answer before you draw anything. | Yes — room types with published densities, and a whole-floor default | Yes — room types with per-type device density | Even |
| Scale calibration Area-based scaling inherits any error in the traced boundary; a known dimension does not. | Click both ends of a known dimension on the plan and type its length | Enter the total floor area of the building | WiFi Heatmap |
Both tools export a PDF. Only one of them puts your name on it.
| Dimension | WiFi Heatmap | Deconflict | Edge |
|---|---|---|---|
| Branding on the report If you are charging for the survey, the document should not advertise the tool you used. | Your logo and company name on the cover | Deconflict's own branding, on every page | WiFi Heatmap |
| Client and site details | Client, site, prepared-by and free-text notes on the cover | Project name and date | WiFi Heatmap |
| Bill of materials The quantity table is the half a quote is built from. | Quantities by model, then a full device schedule | A device schedule, without a quantity roll-up | WiFi Heatmap |
| Method and assumptions This is the page that survives a client, or their IT contractor, asking how the numbers were produced. | An appendix stating the model, the exact attenuation figures used, and what is not simulated | Not included | WiFi Heatmap |
| Per-floor pages with statistics | One page per floor with coverage figures and the signal legend | One page per floor, rendered from the current canvas view | WiFi Heatmap |
| PNG export | Free, watermarked. Unwatermarked on Pro | Free, unwatermarked | Deconflict |
The honest summary: they are free and we are not.
| Dimension | WiFi Heatmap | Deconflict | Edge |
|---|---|---|---|
| Price | Free tier; Pro from €9 for a 7-day pass, €29 a year, or €39 once | Free, with no paid tier | Deconflict |
| Source available If you need to audit or fork the tool, that settles it. | No | Yes — MIT licensed, self-hostable | Deconflict |
| Maintenance A parked project still works. It just will not grow a feature you need, or fix one that breaks. | Actively maintained | Last public commit April 2026 | WiFi Heatmap |
| Support | Email support included with Pro | GitHub issues | WiFi Heatmap |
Both tools will show you a heatmap. The job does not end there. If someone is paying you for the design, what leaves your hands is a document, and that document has your name on it — not ours, and not the tool's.
Cover page
Your logo and company, the client, the site, the date, and the headline coverage figure.
A page per floor
The plan at full resolution with per-floor coverage statistics and the signal legend.
Bill of materials
Quantities by model, then every access point with its band, channel, transmit power, gain and position in metres.
Method and assumptions
The propagation model, the exact attenuation figures used, and an honest list of what is not simulated.
Deconflict exports a PDF too, and it is free. It carries Deconflict's branding, a project name and a device schedule. For your own records that is fine. For a document you are invoicing against, the branding is the problem.
Yes. Deconflict is free with no paid tier, and the source is published under the MIT licence, so you can read it, fork it or self-host it. There is no catch to disclose. WiFi Heatmap has a free tier — five access points, one floor, watermarked PNG export — and charges for the rest.
Five things, concretely. It is free where we charge; it has an access-point placement optimiser and we have none; it traces walls off your floorplan image automatically where we make you draw them; it models channel width per access point (20/40/80/160 MHz) and we plan at 20 MHz; and multi-floor planning plus automatic channel assignment are free there and Pro here. If any of those decide it for you, use Deconflict.
It reports coverage in dBm rather than relative quality bands, so the output can be written into a specification and checked on site. It models antenna gain, which Deconflict does not model at all. Its wall attenuation is tabulated separately for 2.4, 5 and 6 GHz rather than one figure per material. And it exports a client-ready PDF under your own logo, with the client name, a bill of materials with quantities, and an appendix stating the model and its assumptions.
Because that is the unit the rest of the trade uses. A tender that asks for "−67 dBm or better across all occupied areas" cannot be answered with a green square. Anyone can check a dBm prediction with a free phone app after installation; a relative colour band gives them nothing to check against, and gives you nothing to sign off.
Our free tier covers it: five access points is more than a house needs, on one floor, with the full COST-231 engine and no signup — you would only hit the paid tier if you wanted multiple floors or a watermark-free export. Deconflict is also a fine choice here, and free with no cap at all. For a home network with nobody to invoice, the paid deliverables are irrelevant either way; try both, neither asks for an account.
Ours, if you are handing the result to a paying client — the branded report, the bill of materials and the dBm figures are the parts of the job you actually get paid for, and a free tool has little reason to polish them. Use both if you like: they are free to try and neither needs an account.
Comparing something else? WiFi Heatmap vs Hamina · vs NetSpot · every free planner, compared · vs Cambium Wi-Fi Designer
Our free tier gives you five access points on one floor with the full physics engine and a watermarked PNG. That is enough to judge the model against Deconflict's on the same floorplan before anyone pays anything.