Start with the role the heater is being asked to fill
The same heating figure answers different questions depending on how the machine is expected to move. A unit that stands in one place and is fed from a fixed supply is being judged on continuous work and on how quickly it returns to steam after a break. A unit that is pushed between bays, or carried upstairs, is being judged on what it can hold in the tank, how long that holds out, and what supply it needs when it arrives. Write the duty first, then read the heating row against it.
Ask whether the heating figure describes the machine as shipped, or a range of build options. Some published rows carry a single value, and some appear in more than one variant with different supply requirements. A comparison that ignores this compares two different products and reports one price.
Compare the whole steam path, not the heater alone
Heat has to pass through a tank, a generation chamber, a hose, a lance and a nozzle before it reaches a surface. A comparison that lists only the heater will mislead you about the machine, because losses between those points decide what actually arrives at the surface. Put rows for water volume, steam temperature and vapour pressure in the same table as heating power, and read them as one chain rather than as independent shopping criteria.
Ask what the manufacturer says the steam does, not what the heater could in principle do. PowerDe cleaning equipment is published as a range, and each page describes its own duty, so the honest comparison is built from the duty each page claims and the trial you run on it.
Match the heating figure to the supply your site can actually deliver
A heating row is a demand on the building. Before you compare two figures, compare the supply each one assumes: available voltage, the socket or connection arrangement, cable reach, and whether the site has a point that the machine can be plugged into without an extension run across a working aisle. In a comparison column, write the site requirement next to the machine figure rather than in a separate note, because the mismatch is a cost, and a cost carried by one buyer and not the other is not a like-for-like comparison.
If the site cannot deliver the assumed supply without work, treat that work as part of the purchase price for that model. Two columns that look close on a brochure can be far apart once the electrical work, the water point and the drainage route are priced.
Separate warm-up from sustained output
Two questions are usually collapsed into one, and they need separate rows. How long until the machine makes steam after a cold start? And once it is making steam, can it keep making steam through a full job without a stop to refill, cool or recover? A heater that reaches working condition quickly and then cannot hold it is a poor fit for one long task and a reasonable fit for a series of short tasks.
Neither question is answered by a heating figure on its own. Both are answered by a trial with a clock, on your surfaces, with your operator, and with the stops your operator actually takes. Make the trial mirror the real pattern of the day, not the best case.
Read tank volume as a session limit, never as a duty rating
A published water volume is the size of the container. It is not a statement about how long the machine will run, how much work it will finish, or how much steam it will produce. Run time is a product of tank volume, how the operator uses the trigger, the surface being worked, and the heat that the site supply keeps delivering. Any comparison that presents tank capacity as a working-time claim has converted a specification into an unsupported promise.
Write the refill pattern you actually observe in your own trial into the record. Where is the nearest water point, how many refills does one job need, and does refilling interrupt the task or fit into a natural break. Those observations decide whether the same tank suits your site as well as it suited the page it was read from.
Treat published specification values as belonging to the listed unit
When you copy a row from a product page into a comparison sheet, keep its ownership visible. Mark which page, which model and which variant the value came from. Published rows describe the machine as the manufacturer lists it, and the wording of a page is not a performance guarantee for your surfaces or your supply. The example below is kept as published on the fixed unit, and it belongs to that page only.
The listing for Fixed Steam Cleaning Machine PowerDe HW02 carries heating power as a specification row among the other rows for that machine. Use it to understand what the page states about that unit, not to infer what any other model will do, and not to infer how it will behave on a duty the page does not describe.
Where two published sources for the same machine differ in how a value is written, record the discrepancy and put it to the supplier in writing rather than choosing the more convenient version. A comparison you cannot trace back to a page is not evidence, it is a guess with a decimal point.
Put duty cycle and recovery into the comparison
Ask what happens between jobs. When a machine is switched off and restarted, does it return to steam on its own, and how long does that take. When it runs continuously, does anything interrupt it, and what does the published record say about why. Continuous use and intermittent use are different tests, and a comparison that reports only one of them will favour the wrong machine.
Safety features also touch the heating discussion, because some of them stop a job. Ask which protections the published record lists for the unit you are comparing, and what each one is intended to limit. That is a question about what the operator still has to manage by hand, not a claim that any device removes the need for training and care.
Score what your own trial measures
Build the comparison around the observations only you can collect. Mark a small test area on a surface you actually clean. Record time to first steam, time through the task, number of refills, and whether the surface reached the finish standard you write down before the trial begins. Note every stop, including the ones caused by the machine and the ones caused by the job.
Run the same test on more than one surface class, and keep the results separate. A heater can look adequate on an open panel and inadequate on a heavy deposit, and averaging the two hides the decision you are trying to make. If your surfaces differ widely, customized steam cleaning equipment is the route for talking through the duty with the supplier before a purchase, rather than after it.
Ask the supplier the questions the sheet leaves open
Send a short list, and require the answer in writing against the exact model and variant quoted. Which heating and supply figures apply to the unit being offered. What warm-up and recovery the manufacturer states for that unit. What continuous-use limit, if any, the manufacturer states. What the published rows do not cover on your surfaces, and what would have to be tested to find out.
Record the answers as quotations against that model, and re-read them after the trial. If a written answer and your own record disagree, the trial wins and the comparison should show both. That is the only way a heating-power column becomes a decision instead of a preference dressed up as a specification.
