The complaint: Random cracking of 12”x24” glass within days of the installation. The cause: ¾” spots of standard thinset mortar. Glass must be fully embedded in a thin even layer of thinset mortar so undue stress created by normal shrinkage of the mortar is not transferred through the tile.

The solution: Complete replacement.


It was just 12 years ago that I owned and operated a tile contracting business. I spent a good deal of time and money building our brand. We were always looking for the latest trends and products.

We had a training budget that was available to each and every installer to use as they saw fit and rewarded them with a merit increase if they achieved specific educational targets. Comfort levels were developed with customers during the sales process by promoting our trained in-house installation team. We prided ourselves on our work, our equipment, and getting the job done on time. We always took each job personally with a specific individual assigned the responsibility for the installation of each and every job.

While my former business model may appear Darwinian in comparison to today’s current practices, it is evident that many attributes of my former plan, in particular the educational aspects, are still needed and in short supply. Today ceramic tile, stone and glass no longer seem to carry the lofty description of a premium floor or wall product. Instead they have descended to the level of just another floor covering in the minds of most retailers, installers, and end-users.

With this lowered level of perceived value, margins on ceramic and stone products along with labor have been reduced to a more moderate yet still profitable level. Gone is the perception that tile, stone, and glass require specialized knowledge and skills that are uncommon among the majority of floor covering installers.

The complaint: Hollow sounding tile in the school corridors within six months of installation.

The cause: Thinset applied excessively thick (causing shrinkage) over a smooth non-absorbent steel troweled concrete slab.

The solution: The common areas of the school will require complete replacement.

Today’s business climate has forced flooring contractors and retailers alike to take on projects with which they are not totally comfortable, and assign the work to those who are not thoroughly experienced in the knowledge and skills required for bonding a rigid covering to the floor or wall using cement based bonding materials. The results of a failed installation can be much more than a nick in the profit margins. They can prove financially devastating.

The knowledge of both tile products and installation materials required for successful and durable installations has greatly increased in complexity. As a tile inspector and consultant I am often met on a failing job site by a contractor who makes the bellicose statement that he has been doing tile this way for 30 years and never had a problem; obviously it is either defective tile or setting material.

What is sad is that the way they have been doing it for 30 years is indeed usually the problem. Many aspects of ceramic tile products have changed over the years in both the production process and sizes that are now commonplace. Thirty years ago 12” by 12” tile was king and 16” tile was cutting edge.

Glass tile has also experienced many changes. In the early years it was highly unusual to see anything other than the traditional 3/4” tesserae cast glass type product. Now we have multiple methods of manufacturing glass that can produce some very large tile. I was recently on a job with a 2’ by 4’ glass tile! The simple message here is if you are still doing tile installation today the same way you have for 20, 30, or 40 years there is a good chance that you are doing it wrong.

With very little exception there is no bonding material that has not experienced some form of change in recent years. The vast majority of these have been very positive changes. They include accommodation for emerging technology as well as the needs of today’s large format tiles.

The standard thinsets are often unable to provide the required performance attributes for a successful installation. Examples include the need to accommodate minor surface undulations that will require a thicker bed of thinset mortar. For the installer who has been doing it the same way for 30 years, he orders a little more standard thinset anticipating this. Unfortunately, he is likely not aware that a traditional thinset mortar may not support a heavier tile.

He is also likely not aware that a traditional thinset mortar has a maximum profile thickness of 1/4” in height. Once that depth is exceeded, there is a distinct possibility that the thinset will shrink excessively as it cures, in some cases pulling away from the tile and causing bond loss. The likelihood of this is enhanced by the weaker matrix of using a product designed for a 1/4” thickness as opposed to a medium bed mortar, which is designed for a maximum 3/4” thickness.

Preferred sizes of tile products continue to get bigger. Even 15 years ago a 24” tile was somewhat uncommon. With bigger tile there is also the desire for smaller joints. Given the flatness requirements to avoid excessive lippage with bigger tile, the desire for tighter grout joints is almost a contradiction to the realities of installing large tile. If you need both, that requires a very flat floor, one that you are highly unlikely to get without making a specific effort (and expense) to achieve it. While flat floors have always been a concern in the tile trade, historically they could be dealt with at a reasonable cost. With big tile and tight joints that is far from likely in today’s ultra competitive construction environment.

What I see used as an increasingly common strategy around the cost of floor flattening is “dotting,” “spotting “or “pointing” tile with huge gobs of mortar. If I am there to see it, it means it didn’t work. Thin ceramic tile, stone, and glass are not structural products; they are surfacing units. They must be equally supported over their entire surface area. I would guesstimate that not a month has gone by in the last several years where I have not been called to a failing installation and found gobs of mortar supporting the tile in addition to other installation-related maladies. There are no corrective measures for insufficiently supported tile short of replacement.

Limits of space will cause me to skip over the long reigning king of all installation failures: Lack of movement accommodation in the installation. As tile gets bigger, grout joints get smaller and fewer, and the need has never been greater for movement accommodation in tile work. Recently a major upscale hotel chain told me due to increasingly persistent failures over the last three to four years they would no longer use tile unless no other alternative existed. If we don’t start providing some training and education on the needs of tile products and installation, I expect they will be joined by a long list of others.